Baseten on X / Twitter

Official accounts (1)

Baseten @baseten · Company Primary

Active Accounts: 1 tracked
Posts YTD: 352 ↑ vs 209 last year
Avg Posts/Month: 33.1 past 12 months
Views: 19,046,422 397/397 reported
Likes: 21,119 397/397 reported
Reposts: 2,112 397/397 reported
Replies: 1,258 397/397 reported
Quotes: 634 397/397 reported
Bookmarks: 8,570 397/397 reported

Monthly Activity

Account: Date range:

Engagement values are the latest totals observed for each post, grouped by publication month. Provider coverage can vary by metric; unavailable values are shown as --.

Post Details

Search:

Post	Account	Account Type	Post Type	Published	Views	Likes	Reposts	Replies	Quotes	Bookmarks
"With customization, and with partners in the ecosystem that we work with, you could turn these models into the frontie…	@baseten	Company	Original	2026-10-25	2,637	25	3	4	0	4
Meta’s Muse Spark Open is live on the Baseten Loops SDK for post-training. Train it on your data with your reward func…	@baseten	Company	Original	2026-10-25	2,113	31	2	4	3	11
We ran @mirendil's new base model on Baseten's private codebases; that is, code it's never seen. Mirendil-Base-Large ha…	@baseten	Company	Original	2026-10-23	54,131	229	19	9	11	91
With open models, we can build safety tooling directly into the inference layer. Our CEO and co-founder @tuhinone sat…	@baseten	Company	Original	2026-10-23	6,421	71	9	3	1	14
Today we’re announcing the Baseten Environment Network, a group of the best RL environment companies that will help our…	@baseten	Company	Original	2026-10-23	68,635	327	33	18	28	183
.@markchen90, Chief Research Officer at @OpenAI, is joining the All-Gather speaker lineup! This room keeps getting more…	@baseten	Company	Original	2026-10-23	9,229	60	5	2	5	6
Meet Simba 3.5: one model, 48 languages, streaming-native. Compatible with every voice, including cloned voices. Congr…	@baseten	Company	Quote	2026-10-22	1,549	25	3	3	0	2
Thanks, @brexHQ. 💚 We're hiring across the board. https://t.co/q63FBsOgNe https://t.co/0iVcbkQuig	@baseten	Company	Original	2026-10-22	2,558	46	2	1	0	10
Today we’re releasing two decision models, elephant-1 (9B) and mammoth-1 (27B), trained by the @baselabs team, availabl…	@baseten	Company	Original	2026-10-21	33,642	333	38	23	21	166
Agent security belongs in the infrastructure layer. We’re proud to partner with @nvidia on the OpenShell Reference Arc…	@baseten	Company	Quote	2026-10-21	1,551	20	0	4	0	1
The NVIDIA Nemotron Retriever collection of models is now on Baseten! Nemotron 3.5 Embed, Parse, and Rerank help agen…	@baseten	Company	Original	2026-10-21	1,456	27	0	4	0	3
Benchmark performance doesn’t matter if it can’t hold up in production. Major open models launched this month live on …	@baseten	Company	Original	2026-10-21	3,107	50	4	4	0	7
We're excited to bring Mercury 3 to Baseten! - Outperforms GPT-6 Luna, Claude Haiku 5.5 - 1,000+ tokens/sec, 4–8x fast…	@baseten	Company	Quote	2026-10-19	3,415	80	5	22	0	8
How do you build and serve a multi-trillion-parameter model? A deep dive using Kimi K3 as the case study, from Stable…	@baseten	Company	Quote	2026-10-19	8,676	140	1	2	0	119
Proud to support the training and serving for Extend Verify, the first document fraud detection API built for agents. …	@baseten	Company	Quote	2026-10-19	2,338	32	7	5	0	4
Open models now perform close to frontier closed models on coding, at a fraction of the cost. So which tasks are they r…	@baseten	Company	Quote	2026-10-19	2,208	25	2	2	0	3
DeepSeek's latest model, DeepSeek V4.1 Flash 1017, is live on Baseten. At just 763bn parameters, this model delivers…	@baseten	Company	Original	2026-10-17	1,763	28	1	0	0	4
The gathering grows. @bneyshabur, co-founder and CEO of @mirendil, is speaking at All-Gather! We've been overwhelmed (…	@baseten	Company	Original	2026-10-16	1,440	28	4	2	0	4
We are thrilled to serve @reflection_ai's first open-weight model, Beam, on day 0. At 500B parameters, Beam is optimiz…	@baseten	Company	Original	2026-10-16	2,587	59	3	4	0	5
Both of the new PhoneLLM models (30B and 120B, optimized for conversational voice AI) are available as one-click deploy…	@baseten	Company	Quote	2026-10-16	1,162	17	0	7	0	2
We’re excited to serve Muse Spark Open, @Meta's latest open-weight model, on day 0! • 816B-parameter MoE, 42B active…	@baseten	Company	Original	2026-10-15	1,518	40	2	5	0	3
NVIDIA Nemotron 3.5 Super VL is live on Baseten Model APIs and Baseten Loops, day 0! Build and customize multimodal ag…	@baseten	Company	Original	2026-10-15	1,843	34	5	3	0	3
Magicare AI is building the operating system that gets patients into skilled nursing care faster, enabling facilities t…	@baseten	Company	Original	2026-10-14	1,509	22	2	3	2	1
We’re excited to announce @runwayml as a Model Labs partner, and to bring their video generation and editing models to …	@baseten	Company	Original	2026-10-14	6,376	65	5	3	2	7
RT @jaikumarganes: Excited to share that I have joined Baseten. A month ago, one of my former colleagues, who works on…	@baseten	Company	Repost	2026-10-14	141	0	5	0	0	0
Our harness engineering series is now a book. @thealexker and @zakkeener42 cover how to optimize existing harnesses, h…	@baseten	Company	Quote	2026-10-13	3,682	65	4	2	1	46
Gradium TTS by @GradiumAI is now in the Baseten Model Library, pairing ultra-low (50ms) TTFA with natural voices. Use…	@baseten	Company	Original	2026-10-13	3,283	16	1	2	3	2
RT @tuhinone: Safety matters a lot to us at Baseten. As more teams build with AI, we want to give them the signals they…	@baseten	Company	Repost	2026-10-12	113	0	3	0	0	0
RT @tuhinone: agree this is the pattern inside many companies: frontier models for breadth, smaller open models for spe…	@baseten	Company	Repost	2026-10-12	104	0	3	0	0	0
RT @baselabs: New weekly Rollout from the Base Labs team. We ask: why do models forget facts that they iteratively lear…	@baseten	Company	Repost	2026-10-12	116	0	2	0	0	0
AI researchers' idea of a good time: playing HORSE and debating continual learning between shots. @oneill_c (co-head o…	@baseten	Company	Original	2026-10-12	44,958	215	13	12	9	173
RT @GoodfireAI: Goodfire safety infrastructure 🤝 @Baseten inference Today we're partnering with Baseten to bring real-…	@baseten	Company	Repost	2026-10-09	121	0	19	0	0	0
RT @oneill_c: Setting off to build the frontier safety stack for runtime in the open. And we’re already offering it to …	@baseten	Company	Repost	2026-10-09	121	0	5	0	0	0
Open models can come from anywhere. Securing them shouldn’t depend on where you get them. Today we're launching Projec…	@baseten	Company	Original	2026-10-09	17,880	120	18	7	8	30
Congrats to the @_inception_ai team on launching Mercury Decide, the highest-accuracy decision model at the lowest cost…	@baseten	Company	Quote	2026-10-08	3,679	53	5	0	0	12
RT @amiruci: Our first research summit, All-Gather, is on November 18 at the Exploratorium in SF. Apply and see you the…	@baseten	Company	Repost	2026-10-08	111	0	2	0	0	0
We help @NotionHQ build and serve intelligence; then we use it to power our internal knowledge base. While our enginee…	@baseten	Company	Original	2026-10-08	15,918	32	5	8	1	20
Progress that moves the frontier is almost never made alone. We’re thrilled to announce All-Gather, our first-ever res…	@baseten	Company	Original	2026-10-08	64,099	209	19	20	13	95
The Base Labs team will be at NeurIPS this year, presenting 3 workshop papers across continual learning and computer vi…	@baseten	Company	Quote	2026-10-07	2,337	35	2	2	0	6
RT @baselabs: Each week we share ongoing work from across the Base Labs team. We call these releases 'rollouts'. This…	@baseten	Company	Repost	2026-10-05	120	0	10	0	0	0
Our engineers and researchers are constantly experimenting with new methods to improve model performance. Shawn Rushef…	@baseten	Company	Original	2026-10-05	9,716	186	12	12	0	160
"The future is people using the closed frontier, open frontier, and then a long tail of post-trained models that are ta…	@baseten	Company	Quote	2026-10-01	6,359	52	8	3	0	11
The first video-to-video model to pass the Turing test, trained on Baseten. Congrats to the @tavus team on the launch …	@baseten	Company	Quote	2026-10-01	9,899	97	7	5	0	17
When does a speculative decoding draft get accepted? Most theory covers distribution-preserving sampling, but producti…	@baseten	Company	Quote	2026-09-30	3,415	23	1	5	0	7
RT @astuyve: OpenAI's marketplace is the biggest announcement from the day. You can now buy open-weight model inferenc…	@baseten	Company	Repost	2026-09-29	118	0	9	0	0	0
RT @tuhinone: Baseten is now available as one of the first open-model inference providers in the @OpenAI B2B Marketplac…	@baseten	Company	Repost	2026-09-29	105	0	17	0	0	0
RT @amiruci: Two months ago I said the easiest way to use open models in Codex was `brew install baseten-switch`. Start…	@baseten	Company	Repost	2026-09-29	107	0	12	0	0	0
RT @DannieHerz: Baseten is one of the first open model inference providers in OpenAI's new B2B marketplace. This is a h…	@baseten	Company	Repost	2026-09-29	105	0	3	0	0	0
Baseten joined the @OpenAI B2B marketplace today as one of the first open-model inference providers. The best AI compa…	@baseten	Company	Quote	2026-09-29	59,590	171	10	11	10	47
https://t.co/mylK2nCA34	@baseten	Company	Original	2026-09-29	329,742	493	44	16	50	252
RT @baselabs: We're excited to see NVIDIA launch the Open Agent Safety Platform today. We are firmly committed to the…	@baseten	Company	Repost	2026-09-28	119	0	2	0	0	0
RT @amiruci: Our second announcement of the day: Carbon is our newest sandbox infrastructure, built with more capabilit…	@baseten	Company	Repost	2026-09-28	120	0	15	0	0	0
We're deeply committed to building safety controls for the open ecosystem across inference, training, and sandboxes. …	@baseten	Company	Quote	2026-09-28	5,712	39	5	5	1	4
https://t.co/cXemJpcusl	@baseten	Company	Original	2026-09-28	36,432	52	4	1	3	22
RT @DannieHerz: Getting agents into production takes more than better models—it takes an entire ecosystem. We're proud…	@baseten	Company	Repost	2026-09-28	125	0	1	0	0	0
RT @tuhinone: The open frontier only works if we can make autonomy safe by default. @JensenHuang has done important wor…	@baseten	Company	Repost	2026-09-28	118	0	10	0	0	0
We're proud to be a launch partner for the NVIDIA Open Agent Safety Platform and a member of the Open Secure AI Allianc…	@baseten	Company	Quote	2026-09-28	6,094	83	17	9	2	7
Every day, more companies are building with the open frontier. We're thrilled to support them. Congrats @genspark_ai o…	@baseten	Company	Quote	2026-09-26	2,045	6	0	1	0	2
We're officially in our Sheila era. Excited to welcome our new CMO to Baseten! https://t.co/4ExLibgUE2 https://t.co/DH…	@baseten	Company	Original	2026-09-25	31,968	101	8	10	7	15
You can now point an agent at your LangSmith traces, generate a dataset, and start post-training a model with Baseten L…	@baseten	Company	Quote	2026-09-24	2,259	12	1	2	0	6
https://t.co/Kdf1q8VEA1	@baseten	Company	Original	2026-09-24	17,230	42	3	2	2	42
LangSmith Fine-Tuning from @LangChain is in public beta today, and its open-source CLI, smithtune, trains on Baseten Lo…	@baseten	Company	Quote	2026-09-24	4,387	21	1	0	0	4
NVIDIA Nemotron 3 Diarization is available in the Baseten Model Library, day 0! - 500+ concurrent hour-long diarizatio…	@baseten	Company	Original	2026-09-23	1,345	21	1	1	0	3
RL teaches models to work longer, but reasoning is dependent on domain-specific post-training. Baseten's Head of Model…	@baseten	Company	Original	2026-09-22	1,538	25	0	3	0	7
"Your user data, your signal from these models, the improvements to these models themselves will become the core IP of …	@baseten	Company	Quote	2026-09-21	5,918	24	2	0	0	8
We sat down with the team at @elise_ai to talk about why they're betting on specialized, post-trained models, how to ev…	@baseten	Company	Original	2026-09-18	1,330	22	4	8	0	5
We're proud to partner with the team at @p0 to power fast, accurate, low-cost web search for open-weight models via Bas…	@baseten	Company	Quote	2026-09-17	7,300	42	4	3	0	8
We're excited to work with @youdotcom to offer web search with open-weight models at a fraction of the price of closed …	@baseten	Company	Quote	2026-09-17	2,093	18	1	0	0	2
Excited to partner with @KeenableAI to power open-model web search via Baseten Grounded Inference!	@baseten	Company	Quote	2026-09-17	1,517	14	1	0	0	3
Excited to partner with the @ExaAILabs and @ExaDevelopers teams to help us power Baseten Grounded Inference, our new se…	@baseten	Company	Quote	2026-09-17	3,562	29	1	3	1	2
Many of our customers run workloads with speaker attribution, with the strictest demands on both quality and speed. We…	@baseten	Company	Quote	2026-09-17	1,478	18	3	3	0	4
Excited to see Vercel push the open-weights ecosystem forward. 💚	@baseten	Company	Quote	2026-09-17	2,396	16	0	2	0	4
RT @amiruci: Open models on Baseten can now search the web, within the inference path, thanks to our partnership with E…	@baseten	Company	Repost	2026-09-17	111	0	3	0	0	0
Web search is now a first-class citizen for open models, only on Baseten. Baseten Hosted Tools and Baseten Grounded In…	@baseten	Company	Quote	2026-09-16	2,420	22	2	1	0	5
RT @Thom_Wolf: Happy to start collaborating with @baselabs from @baseten and @GoodfireAI to push safety and interpretab…	@baseten	Company	Repost	2026-09-16	105	0	13	0	0	0
Until now, adding web search to open-source models meant hand-wiring orchestration, managing separate keys, and paying …	@baseten	Company	Original	2026-09-16	27,758	98	11	9	7	40
RT @baselabs: We believe openness to be an advantage for AI safety. Openness provides more visibility into the behavior…	@baseten	Company	Repost	2026-09-16	118	0	40	0	0	0
We're proud to support the @LangChain team as they train custom models to outperform the frontier for LangSmith Engine …	@baseten	Company	Quote	2026-09-15	1,970	19	0	1	0	3
https://t.co/MqKH2zI0Mb	@baseten	Company	Original	2026-09-15	22,541	24	1	3	5	11
RT @dwarkesh_sp: Was really interesting to hear John, Beren, and Charlie speculate about why Sonnet 5 and Opus 5 feel l…	@baseten	Company	Repost	2026-09-11	89	0	95	0	0	0
Our own @oneill_c joins @dwarkesh_sp, @johnschulman2, and @BerenMillidge for a deep discussion on long-horizon RL, auto…	@baseten	Company	Quote	2026-09-11	3,277	37	5	5	0	8
DeepSeek V4.1 Flash is live on Baseten Model APIs, day 0: - Smarter, faster, and more efficient than DeepSeek v4 Pro …	@baseten	Company	Original	2026-09-10	2,443	33	2	1	0	5
Thanks Chris Metinko and @axios for covering the story! https://t.co/3wV6cjOh3F	@baseten	Company	Quote	2026-09-10	1,834	15	0	0	0	3
https://t.co/JZXABlmahG	@baseten	Company	Original	2026-09-10	61,723	121	16	9	24	35
A conversation about open-source models in production, plus a Mario Kart tournament. Oct 7 at @Techweek_ in SF by a1…	@baseten	Company	Quote	2026-09-09	2,213	20	1	3	0	2
Speaker diarization is one of the hardest problems in voice AI and the best-performing models can be too slow and expen…	@baseten	Company	Original	2026-09-09	131,170	15	1	3	0	7
We're excited to bring Mercury 2.5 to Baseten: - Sub-200ms p50 latency, great for real-time voice use cases - 1,100+…	@baseten	Company	Quote	2026-09-09	5,774	64	3	3	1	14
We're proud to define the quality-latency Pareto frontier for STT in Coval's benchmarks. Voice AI is an inference prob…	@baseten	Company	Original	2026-09-08	1,422	17	4	3	1	3
AI labs and frontier research teams run RL managed rollouts on Baseten. They publish new weights, and we pick them up a…	@baseten	Company	Quote	2026-09-08	1,753	29	1	1	0	6
We partnered with @harvey to train agents for M&A diligence and showed that model-harness co-optimization can improve a…	@baseten	Company	Quote	2026-09-08	1,548	34	3	1	0	10
RT @philipkiely: Tomorrow at 11 AM PT, I'll be live on air with Declan Jackson of Artificial Analysis to discuss GLM-5.…	@baseten	Company	Repost	2026-09-07	110	0	7	0	0	0
GLM-5.3 is now live with vision, only on Baseten Model APIs. Generate and convert images to code thanks to our post-tr…	@baseten	Company	Original	2026-09-04	2,585	50	0	2	0	8
RT @DannieHerz: Notion helps us think. We help them listen. Notion's AI Meeting Notes serve countless teams (ours inclu…	@baseten	Company	Repost	2026-09-04	111	0	2	0	0	0
Notion's AI Meeting Notes run on Baseten, and Baseten's knowledge base runs on Notion. Our teams have been working toge…	@baseten	Company	Original	2026-09-04	25,159	98	10	8	7	28
"I've been through the lifecycle of building a product that people love. And I should be using that information to make…	@baseten	Company	Quote	2026-09-04	2,214	22	0	0	0	4
We're proud to support companies like OpenEvidence that are pushing the frontier of AI in their specialized domains. T…	@baseten	Company	Quote	2026-09-04	5,734	65	0	2	0	14
RT @DannieHerz: Baseten was built on the belief that the health of the open-source AI ecosystem matters. Base Labs is o…	@baseten	Company	Repost	2026-09-03	129	0	2	0	0	0
RT @tuhinone: We're very excited to launch Base Labs to democratize open-source AI research. We'll push the frontier i…	@baseten	Company	Repost	2026-09-03	107	0	11	0	0	0
RT @amiruci: We have clearly benefited a ton from the ecosystem of ever-improving open models and the published researc…	@baseten	Company	Repost	2026-09-03	103	0	5	0	0	0
We’re excited to keep pushing the open-source AI ecosystem forward.	@baseten	Company	Quote	2026-09-03	14,696	165	8	2	0	51
We released a remote MCP server and skill so coding agents can operate the Baseten platform faster and more efficiently…	@baseten	Company	Quote	2026-09-03	4,310	61	2	2	1	13
Today, we're releasing GLM-5.3 Fast: one of the most intelligent open-weight models ever at an even higher TPS. Design…	@baseten	Company	Original	2026-09-03	17,155	210	11	13	1	59
RT @LangChain: Happening today! An evening of technical talks, meeting builders, and end-of-summer vibes with @PrimeIn…	@baseten	Company	Repost	2026-09-02	95	0	3	0	0	0
RT @philipkiely: https://t.co/ESeIBZP3zR	@baseten	Company	Repost	2026-09-02	116	0	43	0	0	0
We're excited to sponsor PACT (Pressure-Applied Compliance Testing), a new benchmark for enterprise assistant rule-foll…	@baseten	Company	Quote	2026-09-01	3,314	43	1	5	0	8
We're honored to be named a 2026 IA40 winner! The IA40 recognizes the 40 most important private companies in applied A…	@baseten	Company	Original	2026-09-01	2,771	73	2	4	0	6
RT @philipkiely: Teach your agents inference. The full text of Inference Engineering is now available to LLMs everywh…	@baseten	Company	Repost	2026-08-31	120	0	60	0	0	0
RT @LangChain: 🌉 Happening 9/2 in SF! Join LangChain, @baseten, and @primeintellect for a dive into continual learning…	@baseten	Company	Repost	2026-08-31	109	0	3	0	0	0
"If you're not taking care of yourself, you're not writing good code." We were honored to host @bryan_johnson and @sar…	@baseten	Company	Quote	2026-08-31	3,603	29	0	2	0	2
RT @amiruci: Even hackernews is raving about this model!	@baseten	Company	Repost	2026-08-29	118	0	1	0	0	0
Our kernel engineers built an agentic framework to automatically find, build, validate, and ship optimized kernels into…	@baseten	Company	Quote	2026-08-29	13,406	140	5	3	1	91
RT @thealexker: https://t.co/YPKWwqEQ3F	@baseten	Company	Repost	2026-08-28	121	0	50	0	0	0
RT @tuhinone: GLM 5.3 is amazing, the rate of progress of open-weights is clearly continuing to accelerate, and we'll h…	@baseten	Company	Repost	2026-08-28	108	0	5	0	0	0
GLM-5.3 is live on Baseten Model APIs, day 0. - The smartest open-weight model at 743B params - 1M token context - US …	@baseten	Company	Original	2026-08-28	15,660	64	5	4	7	18
RT @philipkiely: Should you use GLM-5.3 or GLM-5.3-Flash (Ox Alpha)? - 753B/40B vs 320B/18B params - 60 vs 57 AA score…	@baseten	Company	Repost	2026-08-28	107	0	4	0	0	0
Post-train GLM-5.3 and GLM-5.3-Flash on Baseten Loops. Inference + training support on day 0. https://t.co/mbIxzhP7l…	@baseten	Company	Original	2026-08-28	7,635	81	5	6	3	28
We're proud to be the fastest inference provider on Artificial Analysis, OpenRouter, and Hugging Face for GLM-5.3-Flash…	@baseten	Company	Original	2026-08-27	21,510	210	9	13	2	45
👀	@baseten	Company	Quote	2026-08-27	5,885	77	0	2	0	4
GLM-5.3-Flash is live, day 0, on Baseten Model APIs. - Useful for general intelligence + agentic coding - More intelli…	@baseten	Company	Original	2026-08-26	9,162	123	4	9	2	21
RT @thealexker: glm-5.3-flash is a story of efficiency. top gems: > outperforms glm-5.2 across domains at 1/10 th…	@baseten	Company	Repost	2026-08-26	116	0	5	0	0	0
RT @amiruci: GLM 5.3 Flash post-training is supported in Baseten Loops on day 0. And there's good reason for it. DM @on…	@baseten	Company	Repost	2026-08-26	102	0	6	0	0	0
Our engineers are cooking. Stay tuned.	@baseten	Company	Quote	2026-08-26	9,387	193	3	8	2	4
RT @oneill_c: If you want to RL (or sft/opd) glm-5.3-flash, DM me to get access to it on @baseten loops. We're very exc…	@baseten	Company	Repost	2026-08-26	106	0	17	0	0	0
RT @gambhir_amit: https://t.co/WCLrBFAHQX	@baseten	Company	Repost	2026-08-25	117	0	4	0	0	0
RT @bryan_johnson: Doing a private longevity event with @baseten in SF on Wednesday. > biological age testing &gt…	@baseten	Company	Repost	2026-08-24	100	0	11	0	0	0
RT @saranormous: Doing a private conversation with @bryan_johnson on longevity + tech on Aug 26 (with bioage testing on…	@baseten	Company	Repost	2026-08-21	92	0	8	0	0	0
Some questions demand the most accurate answers. @youdotcom built its Answer API to help users get grounded, cited answ…	@baseten	Company	Original	2026-08-21	6,623	33	8	3	2	9
GLM-5.2 now supports vision in production, only on Baseten. Ingest images and convert them to code, thanks to our post…	@baseten	Company	Original	2026-08-21	25,654	65	0	5	4	17
More evidence of a many-model future. Congrats @OpenRouter!	@baseten	Company	Quote	2026-08-20	2,690	27	2	0	0	2
RT @hwchase17: top tier webinar tmrw! @Vtrivedy10 (@LangChain) @willcb (@PrimeIntellect) @AEllisBloor (@baseten) and I…	@baseten	Company	Repost	2026-08-20	107	0	14	0	0	0
Last week, the entire Baseten team came together in Chicago for our biannual company offsite. Since the beginning, we'…	@baseten	Company	Original	2026-08-19	4,114	77	9	1	3	11
RT @amiruci: DeepSeek-V4-Pro-0813 came out a few days ago. Baseten is the fastest provider according to @OpenRouter, @A…	@baseten	Company	Repost	2026-08-18	97	0	3	0	0	0
RT @p0: Next Thursday, August 20th, @baseten is hosting @paraga for a live discussion about how agents are changing the…	@baseten	Company	Repost	2026-08-18	98	0	1	0	0	0
We're proud to be the fastest inference provider on Artificial Analysis for DeepSeek V4 Pro 0813, at 147 TPS. Our eng…	@baseten	Company	Original	2026-08-18	11,792	69	1	3	1	20
Inference Engineering is free for everyone to read: https://t.co/sF1pX4LSQB	@baseten	Company	Quote	2026-08-18	167,562	1,853	125	26	11	2,514
Qwen3.8-27B ranks #7 on Artificial Analysis' Agentic Index at only 27B parameters, a fraction of the size (and cost) of…	@baseten	Company	Quote	2026-08-18	2,654	22	1	2	0	7
We are big fans of @NotionHQ. Proud to power the inference behind the product our own team can't work without.	@baseten	Company	Quote	2026-08-18	4,963	20	0	2	0	6
RT @julien_c: Own your models https://t.co/qOVTxeflTk	@baseten	Company	Repost	2026-08-17	112	0	10	0	0	0
Run SFT or RL on Qwen3.8-27B, the most intelligent small model to date. Day 0 support on the Baseten Loops SDK: https:…	@baseten	Company	Original	2026-08-15	4,319	40	5	0	1	11
DeepSeek V4 Pro 0813 is live day 0 on Baseten Model APIs. - Frontier-level Artificial Analysis Intelligence score - Ze…	@baseten	Company	Original	2026-08-14	910,196	73	0	5	1	15
RT @philipkiely: https://t.co/fg4jUePnls	@baseten	Company	Repost	2026-08-14	102	0	7	0	0	0
RT @DannieHerz: Bryan Johnson is truly an obsessive. On August 26th, he'll join us at our HQ for dinner, a private conv…	@baseten	Company	Repost	2026-08-13	117	0	2	0	0	0
At Baseten, we have a thesis: obsessives move the world forward. Few people embody that more completely than @bryan_joh…	@baseten	Company	Original	2026-08-13	132,231	134	5	9	5	27
Coming to Baseten Model APIs today. https://t.co/kUNeDbdqx1	@baseten	Company	Original	2026-08-13	2,752	46	0	2	1	10
We are strong believers in a healthy frontier ecosystem and are dedicated to building the inference and training to sup…	@baseten	Company	Quote	2026-08-11	3,916	22	0	2	0	3
A clear example of how fine-tuning a model on your unique data can yield a much cheaper, specialized model. Thanks to …	@baseten	Company	Quote	2026-08-11	2,917	18	1	1	0	9
NVIDIA Nemotron 3.5 Lightning is live on Baseten day 0! This is the fastest open model in its class, built for long-ru…	@baseten	Company	Original	2026-08-11	3,079	50	1	3	1	8
Muse Glimmer, Meta's new open-weight agentic model, is available on Baseten day 0! - 30B parameters, including a 1.8B …	@baseten	Company	Original	2026-08-11	3,216	44	1	4	0	4
We're proud to be @arcprize's inference partner as they test the impact of models like DeepSeek V4 Flash and Kimi K3. …	@baseten	Company	Quote	2026-08-07	3,878	35	2	1	0	5
Baseten is now an official inference provider on @huggingface 🤗 Run Kimi K3, DeepSeek V4 Flash, and GLM-5.2 on Baseten…	@baseten	Company	Original	2026-08-06	50,640	192	15	11	5	26
We often see smaller, specialized models outperform the frontier on specific tasks. Fine-tuning DeepSeek V4 Flash achie…	@baseten	Company	Original	2026-08-05	3,856	53	6	2	1	19
"A really strong open-weight ecosystem will lead to an even stronger closed-weight ecosystem. It's necessary for both p…	@baseten	Company	Original	2026-08-05	2,821	37	3	7	0	11
Open-source models have reached a new level of intelligence; our own engineers use models like GLM-5.2 for tasks like k…	@baseten	Company	Quote	2026-08-04	5,016	51	1	2	0	12
One of our engineers asked @poolsideai's Laguna S 2.1 to transform a 715-file C++ game from neon cyberpunk into an Anci…	@baseten	Company	Original	2026-08-03	3,541	33	2	4	0	4
Yes, we will have Qwen3.8-Max.	@baseten	Company	Quote	2026-08-03	5,684	109	3	8	0	1
Try DeepSeek V4 Flash on our Model APIs today. - 80–98% cheaper than other frontier models - Comparable intelligence -…	@baseten	Company	Original	2026-08-01	23,232	139	5	9	4	21
Qwen3-TTS supervised fine-tuning (SFT) for voice cloning is now available on Baseten Training. We see 16% faster time …	@baseten	Company	Quote	2026-07-31	3,841	20	2	1	0	8
Thinking Machines Lab’s new Inkling-Small is now available on our Model APIs! At 276B parameters (12B active), Inkling…	@baseten	Company	Original	2026-07-30	2,807	45	2	3	0	4
The easiest way to run open (Kimi K3, GLM-5.2) and closed models in your harness: Baseten Switch!	@baseten	Company	Quote	2026-07-30	4,056	56	3	1	0	17
Today, we're announcing Baseten for Model Labs. We believe the AI landscape will be made up of a diverse ecosystem of …	@baseten	Company	Original	2026-07-29	49,081	219	25	11	19	74
Thrilled to be partnering with companies like @harvey that value research as much as we do.	@baseten	Company	Quote	2026-07-29	4,188	37	1	2	0	4
We shipped 18x faster tokenization for Kimi K3 on day 0. Michael Feil is such a cracked engineer, we even let him nam…	@baseten	Company	Quote	2026-07-29	7,931	73	1	1	0	18
Proud to support the Cursor team with day-0 support for Kimi K3!	@baseten	Company	Quote	2026-07-28	13,898	129	1	1	0	10
Happening today! Link to register in thread.	@baseten	Company	Quote	2026-07-28	3,805	10	1	1	0	1
"The central change is not scale alone. Each architectural step changes what the model stores, how it updates that stat…	@baseten	Company	Quote	2026-07-28	7,191	66	4	1	1	19
Tomorrow we're hosting a virtual discussion and Q&A on how to evaluate Kimi K3 for enterprise workflows. Join our …	@baseten	Company	Original	2026-07-27	11,658	29	1	2	3	6
Serving a 2.8T parameter frontier model like Kimi K3 at scale, on day 0, takes a village.	@baseten	Company	Quote	2026-07-27	5,334	90	1	4	0	11
Kimi K3 is now live on our Model APIs, day 0. https://t.co/Itz5SDiEYz	@baseten	Company	Original	2026-07-27	684,174	612	48	33	28	111
Jeanne DeWitt Grosser, the COO of Vercel, took a 10-person SDR team down to one person and an AI agent. At Stripe, she…	@baseten	Company	Original	2026-07-24	3,197	15	1	2	1	5
Kimi K3 drops Monday, and our team is already cooking. Get on the Kimi K3 drop list and get $25 of free Model API cred…	@baseten	Company	Original	2026-07-24	56,897	811	26	16	6	201
You can fine-tune GLM-5.2 on the Baseten Loops SDK and easily deploy your checkpoint to a production inference endpoint…	@baseten	Company	Original	2026-07-24	4,424	88	6	4	0	16
Today, we're introducing GLM-5.2 Fast: our GLM-5.2 Model API designed for the most demanding real-time use cases. The …	@baseten	Company	Original	2026-07-23	114,682	538	29	27	14	169
We added vision support to GLM-5.2 with just a 2-layer MLP, <50M parameters. Shout out to @part_harry_ for his wor…	@baseten	Company	Quote	2026-07-23	5,669	70	4	0	0	9
In case you missed it: @part_harry_ didn't just add vision support to GLM-5.2, he also made it open source. We love…	@baseten	Company	Quote	2026-07-23	3,196	33	0	2	0	4
The future will be built on many models, open and closed, general and specialized at different levels of granularity. …	@baseten	Company	Quote	2026-07-22	1,949	19	0	1	0	2
The American open-weight ecosystem is growing. We're thrilled to support Poolside as they push open-weight coding model…	@baseten	Company	Quote	2026-07-21	4,248	39	3	1	0	5
You can now generate 5 seconds of video in under 2.5 seconds, courtesy of the Baseten kernels team.	@baseten	Company	Quote	2026-07-20	3,773	25	0	0	0	3
"So, can a model learn facts continually in its weights? Creating usable knowledge: solvable Preserving capability: so…	@baseten	Company	Quote	2026-07-17	5,175	27	1	3	0	14
We're building in Canada! We opened two new offices in Toronto and Montreal, and we're growing fast. If you’re excited…	@baseten	Company	Original	2026-07-16	41,424	147	17	13	7	45
NVIDIA Nemotron 3 Embed 8B and 1B just dropped, and we're excited to offer them day 0 in our Model Library! We're also…	@baseten	Company	Original	2026-07-16	1,907	27	1	1	0	3
GLM 5.2 is one of the best open models available, but it can't support image inputs natively. Until now.	@baseten	Company	Quote	2026-07-16	6,660	42	0	2	0	11
Inkling is live on Baseten. We’re proud to partner with @thinkymachines to provide day 0 support. Try it here: https:/…	@baseten	Company	Original	2026-07-15	4,702,745	87	6	10	4	17
Join us for Built on Baseten: a showcase for early-stage founders and technical builders who are owning their intellige…	@baseten	Company	Original	2026-07-15	1,616	14	1	4	0	4
Some workloads demand either the highest throughput or the lowest latency. Embedding workloads need both. We built Bas…	@baseten	Company	Quote	2026-07-14	1,984	17	1	0	0	2
Research has always been core to Baseten, from training to model performance and beyond. Charlie, who leads part of o…	@baseten	Company	Quote	2026-07-14	3,321	25	2	1	0	6
Step 3.7 Flash is now in the Baseten Model Library! This is a 198B-parameter sparse MoE model (11B active per token) w…	@baseten	Company	Original	2026-07-14	7,899	17	0	2	1	2
We're seeing it too: companies of all sizes are adopting more open-source AI, including alongside closed models. The f…	@baseten	Company	Quote	2026-07-14	4,555	31	0	2	0	6
Agents will use the web more than humans ever have. High-accuracy AI web search is what @p0 is built for. We're proud …	@baseten	Company	Quote	2026-07-10	7,481	46	5	6	0	19
https://t.co/rVIxR6TkDK	@baseten	Company	Original	2026-07-10	18,080	48	3	3	4	29
Excited to be featured in Notion's newest customer story on how we're scaling expertise with Custom Agents! Meet katzg…	@baseten	Company	Quote	2026-07-08	4,473	36	0	3	0	6
If you're at RAISE Paris today, come see @tuhinone on the master stage at 4:40 p.m. with @CorinneMRiley! https://t.co/c…	@baseten	Company	Original	2026-07-08	3,612	17	3	2	1	1
Thank you, @tbpn, for having our CEO @tuhinone on to talk about how companies of all sizes are investing in owning thei…	@baseten	Company	Original	2026-07-07	6,684	57	2	6	0	22
We partnered with Braintrust to evaluate how GLM-5.2 performs on long-context retrieval tasks. Not only is its retrie…	@baseten	Company	Quote	2026-07-07	5,849	44	1	3	0	19
The future will be built on many models. Open-source: good enough for many use cases at a small fraction of the price.…	@baseten	Company	Quote	2026-07-07	9,768	62	2	7	1	24
The future will be built on many models, each intentionally chosen for specific and specialized tasks. Great read fro…	@baseten	Company	Quote	2026-07-06	8,793	39	1	1	0	23
We're thrilled to power Laguna XS 2.1 from Poolside via the Baseten Frontier Gateway! Laguna XS 2.1 supports a 256K co…	@baseten	Company	Quote	2026-07-02	4,097	20	1	2	0	2
Catch our CEO @tuhinone live on @tbpn at 1 p.m. PT! https://t.co/cCuTfB58X8	@baseten	Company	Original	2026-07-02	1,288	21	2	2	0	3
We made an early bet on multi-cloud infrastructure. Our Multi-cloud Capacity Manager (MCM) abstracts 18+ clouds, which:…	@baseten	Company	Original	2026-07-01	1,571	20	2	2	1	3
We partnered with Braintrust to highlight the real-world strength of the latest open-source model. For long-context re…	@baseten	Company	Quote	2026-06-30	6,436	52	4	1	0	27
Plus: - You own the performance optimizations. - You can specialize them for your exact use case. We love open-source!	@baseten	Company	Quote	2026-06-30	3,824	43	3	2	0	5
Speechify was built to make learning accessible to everyone. Simba 3.0 powers its voice-first productivity platform for…	@baseten	Company	Original	2026-06-29	7,138	36	10	2	3	9
Voice AI is unforgiving: any delay ruins the user experience. We're honored to be SpeechifyAI's sole provider powering …	@baseten	Company	Quote	2026-06-29	2,893	26	4	2	0	3
NYC team repping at Nasdaq tower this week! 💚 We're growing fast, and we're hiring across the board. If you're excited…	@baseten	Company	Original	2026-06-26	4,050	73	5	7	2	10
Live draft model training is now part of our Speculation Engine in the Baseten Inference Stack. Where rolled out, we s…	@baseten	Company	Quote	2026-06-26	3,879	32	0	3	0	12
Excited to power GLM-5.2 on @cline! How to use it in about 10 seconds: https://t.co/MOrWq9qtda	@baseten	Company	Original	2026-06-25	8,892	86	7	4	1	18
"Frontier models for the hardest general intelligence and post-trained open source for high-volume and specialized work…	@baseten	Company	Quote	2026-06-24	4,571	42	1	5	0	12
You can now access our GLM-5.2 API through the Merge Gateway! GLM-5.2 matches frontier model intelligence while runnin…	@baseten	Company	Original	2026-06-24	4,716	90	10	9	0	19
"That's when they come to open-source models, that's when they come to Baseten, that's when they come to post-train mod…	@baseten	Company	Quote	2026-06-23	4,774	41	8	3	1	7
Excited to be a day 0 launch partner for BioNeMo, NVIDIA's new, fully-open agent toolkit for scientific workflows! Al…	@baseten	Company	Quote	2026-06-23	4,952	42	5	4	0	5
The best open model with the best performance: GLM-5.2 runs at >280 TPS and <0.8s TTFT on Baseten. Try it here: …	@baseten	Company	Quote	2026-06-22	30,442	129	5	12	6	46
We’re excited to announce our $1.5B Series F. Baseten exists to help companies own their intelligence and run AI produ…	@baseten	Company	Quote	2026-06-22	95,001	280	23	28	10	70
Most supervised fine-tuning (SFT) studies run on generic data. Ours run on production tasks, paired with evals our team…	@baseten	Company	Quote	2026-06-18	3,544	20	2	3	0	10
Moving from closed to open models can cut costs significantly. See how much you can save by using GLM-5.2: https://t.c…	@baseten	Company	Quote	2026-06-17	3,404	25	0	4	0	7
Ever wonder how much money you'd save by switching to open-source models? We just launched a cost calculator for that…	@baseten	Company	Original	2026-06-17	5,816,328	230	14	20	0	30
Kimi K2.7 Code is live on Baseten Model APIs. Kimi K2.7 Code has 30% lower reasoning-token usage compared to K2.6, wit…	@baseten	Company	Original	2026-06-17	910,406	58	4	6	1	6
We 💚 @cursor_ai, thanks for inviting some of our post-training experts to give chalk talks at Cursor Compile!	@baseten	Company	Quote	2026-06-17	2,376	20	2	2	0	5
Baseten is proud to partner with the Boltz team on this incredible launch.	@baseten	Company	Quote	2026-06-16	4,221	31	3	0	0	8
I spy ... GLM 5.2 on @NotionHQ! Check out https://t.co/e4br5JXagu’s latest advanced agentic and reasoning model, power…	@baseten	Company	Original	2026-06-16	2,981	36	1	1	0	6
GLM 5.2 is live on Baseten. 5.2 is built for agentic engineering: stronger coding, sharper agentic reasoning, and a lo…	@baseten	Company	Original	2026-06-16	1,927,023	141	8	8	5	33
The new AgentPerf benchmark by @ArtificialAnlys shows that @NVIDIAAI Blackwell delivers best performance for demanding …	@baseten	Company	Original	2026-06-12	2,476	16	3	0	1	5
We're thrilled to be working with the Harvey team to push open models to frontier-level performance for legal AI. Sho…	@baseten	Company	Quote	2026-06-12	2,578	20	2	3	0	3
Congrats to the MiniMax team on the open-source launch of M3! There are very few <500bn parameter models that can t…	@baseten	Company	Original	2026-06-12	21,638	43	3	5	1	10
Join Baseten, Lovable, and ElevenLabs to hack on the future of healthcare.	@baseten	Company	Quote	2026-06-12	932	9	0	2	0	1
We've heard from customers that they ship model updates >50% more often with rolling deploys than their previous sol…	@baseten	Company	Quote	2026-06-12	3,450	23	1	3	0	15
Great to see @Baseten’s own @oneill_c and @part_harry_ sitting down with @cursor_ai’s @sjwhitmore to talk about the man…	@baseten	Company	Quote	2026-06-11	3,182	16	1	3	0	5
We are excited to announce that we have partnered with @_inception_ai to make Mercury 2 available on Baseten. This mak…	@baseten	Company	Quote	2026-06-11	26,337	46	5	3	5	12
https://t.co/HGVpRPEhBo	@baseten	Company	Original	2026-06-11	34,159	43	6	2	10	11
The longer the context, the more memory your LLM needs. We introduce research techniques to compress that memory 200x o…	@baseten	Company	Quote	2026-06-10	3,666	39	0	5	1	9
Baseten is live on the Respan Gateway. Congratulations to the @RespanAI team on their Gateway launch as they bring obs…	@baseten	Company	Original	2026-06-09	1,324	14	7	3	0	2
Join Charlie for a conversation with @thatsjonsense and @sarahmsachs on how @GammaApp and @NotionHQ think about model s…	@baseten	Company	Quote	2026-06-08	2,663	15	0	4	0	1
GLM 5.1 now achieves 160+ TPS and <2-second TTFT on Baseten. Ideal for agentic workloads that need high throughput …	@baseten	Company	Original	2026-06-05	6,909	91	2	8	1	15
Are you tired of waiting 17 minutes for an AI agent to finish a code change? As an agent’s context grows, standard tra…	@baseten	Company	Quote	2026-06-04	1,517	20	0	2	0	3
We are excited to welcome Gabe Stern as General Counsel. Welcome, Gabe! https://t.co/fSpxLhZpCx	@baseten	Company	Original	2026-06-04	7,526	14	0	2	2	2
Agents append to their own context. But attention is quadratic, so 2x context = 4x work per step. Nemotron 3 Ultra swa…	@baseten	Company	Quote	2026-06-04	1,986	32	4	3	1	13
10M developers use @opencode every month. This means the experience has to feel the same every hour of every day; slow …	@baseten	Company	Original	2026-06-03	14,266	98	4	6	0	23
Today, Baseten and @MicrosoftAI are excited to announce that MAI-Thinking-1 is coming to Baseten. MAI-Thinking-1 is a …	@baseten	Company	Original	2026-06-02	15,919	65	5	3	2	18
We're hosting an Inference Café for a16z tech week NYC at Verci Flatiron! For the folks experimenting with frontier op…	@baseten	Company	Original	2026-06-01	2,763	27	1	1	0	6
Philip worked with @rimelabs to create an enterprise-grade voice clone for an ambitious project: an AI-narrated audiobo…	@baseten	Company	Quote	2026-06-01	3,133	21	2	2	1	5
Proud to partner with JetBrains. Congrats on the launch!	@baseten	Company	Quote	2026-06-01	2,681	23	0	1	0	1
Want to teach a robot to open a door without shattering the glass, crushing the handle, or clipping through reality? If…	@baseten	Company	Original	2026-06-01	1,984	23	0	5	0	9
We’re honored to be featured in the @Redpoint Infrared 100 list alongside our incredible customers. As we enter the mu…	@baseten	Company	Quote	2026-05-29	3,328	28	2	4	0	2
Our kernels team shipped 2.5x faster FLUX.2 image gen without visible quality loss. How they did it: Distribution Matc…	@baseten	Company	Quote	2026-05-29	1,697	16	0	3	0	2
Our research team partnered with @Harvey and showed that post-trained open models can compete at the frontier on LAB, t…	@baseten	Company	Quote	2026-05-27	6,043	37	6	4	0	18
Excited to support the @trajectorylabs team as their inference partner. Powering live, continuous post-training for f…	@baseten	Company	Quote	2026-05-27	5,925	42	4	4	1	3
Most enterprise AI models stay static after deployment, forcing teams to rely on endless prompt tweaking. This is the …	@baseten	Company	Original	2026-05-27	7,268	92	4	12	0	27
If you haven’t built AI automation that relies on documents, it’s easy to underestimate how hard it is. Parsing perform…	@baseten	Company	Quote	2026-05-26	2,830	21	1	5	0	3
Biotech R&D is generating more scientific AI models than ever, from protein structure prediction to molecular docking t…	@baseten	Company	Original	2026-05-20	2,313	35	10	1	0	10
“Intensity plus joy — an aha moment for me at Baseten is that those two things are not on opposite sides of a spectrum.…	@baseten	Company	Quote	2026-05-19	11,264	24	1	0	1	8
Sub-second image generation with Flux.2 [dev] and Qwen-Image: Flux.2 [dev]: 2.3x faster, 0.98s latency (B200) Qwen-Ima…	@baseten	Company	Quote	2026-05-18	3,324	24	2	1	1	5
Last week, we launched Baseten Frontier Gateway. This week, Marylise sits down with Bola to talk about why.	@baseten	Company	Quote	2026-05-15	4,247	19	0	1	0	3
We serve Qwen3-TTS on vLLM-Omni at $3 per 1M characters. That's 90% lower in cost than comparable closed-source TTS API…	@baseten	Company	Quote	2026-05-14	76,017	97	8	3	5	80
“The question every app layer company is now asking is no longer ‘how do we use AI?’ It is ‘how do we resist commodific…	@baseten	Company	Quote	2026-05-13	1,865	21	0	0	0	12
Intelligence should be defined by the people closest to the work. Intelligence should be owned by all of us. Let’s bui…	@baseten	Company	Quote	2026-05-13	35,848	47	4	3	0	20
We're proud to share our partnership story with @SpeechifyAI. Speechify just announced SIMBA 3.0, ranked top 10 global…	@baseten	Company	Quote	2026-05-12	1,676	16	7	1	1	0
https://t.co/D0Oo41Bgnn	@baseten	Company	Original	2026-05-12	5,225	23	2	0	3	6
EliseAI is the leading AI startup automating complex housing and healthcare systems, which are two of the most operatio…	@baseten	Company	Original	2026-05-11	4,693	28	3	0	1	13
DFlash vs EAGLE on a single B200 with Qwen3-8B: EAGLE: ~2x speedup, autoregressive drafting (one forward pass per toke…	@baseten	Company	Quote	2026-05-08	44,163	160	13	8	4	133
Open-source RL libraries break at frontier scale. We built Baseten Loops to fix this. Loops is a training SDK that tak…	@baseten	Company	Quote	2026-05-08	67,291	110	7	2	6	110
We are excited to share our partnership story with @poolsideai and proud of what we've built together. Poolside has be…	@baseten	Company	Original	2026-05-07	1,808	37	1	0	0	7
"No post-training Pre-PMF" Our CEO and Co-Founder @tuhinone sat down w/ @saranormous and @eladgil on @nopriorspod to d…	@baseten	Company	Quote	2026-05-06	9,080	36	4	1	0	20
https://t.co/sGB4D3iKel	@baseten	Company	Original	2026-05-06	24,370	59	9	3	9	27
Harvey is raising the bar in legal AI. Congratulations on launching the Legal Agent Benchmark! Baseten is proud to supp…	@baseten	Company	Quote	2026-05-06	2,007	25	2	0	0	2
Introducing @NVIDIA Nemotron 3 Nano Omni. NVIDIA Nemotron 3 Nano Omni is an open multimodal foundation model that unif…	@baseten	Company	Original	2026-04-28	4,968	32	8	1	1	5
Laguna XS.2 has landed, and it’s live on Baseten. We’ve baked in inference optimizations so you can deploy @poolsideai'…	@baseten	Company	Original	2026-04-28	3,833	52	7	2	0	4
The strongest open-source agentic model is live on Baseten! DeepSeek V4 is a preview of two powerful MoE models: V4-Pr…	@baseten	Company	Original	2026-04-24	3,701	70	2	4	0	14
We made an early bet that our permissions model needed to handle complex, many-to-many relationships, and orgs of any s…	@baseten	Company	Quote	2026-04-23	3,023	13	1	0	0	7
Our engineers developed RadixMLP, a technique that exploits the position-wise nature of MLPs, LayerNorms, linear projec…	@baseten	Company	Quote	2026-04-22	13,308	86	6	0	2	77
Kimi K2.6 has landed, and it is live on Baseten! We have baked in multiple inference optimizations so that you can lev…	@baseten	Company	Original	2026-04-20	116,549	141	8	12	4	46
Most people installing AI coding agents are optimizing the wrong things. Harnesses are the new IDE. Here are the three…	@baseten	Company	Quote	2026-04-17	5,788	36	1	1	0	56
Honored to be featured in the Forbes 2026 AI 50 List alongside so many of our great customers! https://t.co/gMsw9PrPM…	@baseten	Company	Original	2026-04-16	1,801	24	6	1	0	6
Speculative decoding is one of the highest-leverage inference optimizations out there. Here are tactical steps on how …	@baseten	Company	Quote	2026-04-13	7,414	66	3	0	0	58
HumanX was a blast! Some highlights: -> @tuhinone and @saranormous took the main stage to discuss what it takes to …	@baseten	Company	Original	2026-04-13	1,311	11	1	0	0	0
We recently launched BDN for 2–3x faster cold starts for large models at massive scale. This is how it works.	@baseten	Company	Quote	2026-04-09	4,996	31	1	0	1	25
Come jam on open-source agents with us! We're hosting a NYC hackathon with our friends at @veris_ai on April 18 in SoH…	@baseten	Company	Original	2026-04-09	1,232	13	1	1	1	4
In finance, speed is a competitive edge. @hebbia powers the AI that gives the world's leading financial institutions b…	@baseten	Company	Original	2026-04-08	7,092	44	7	3	3	17
We're at HumanX all week! Here's what to expect today: -> Lightning talk at the Vultr booth with @philipkiely (3 PM) …	@baseten	Company	Original	2026-04-07	812	12	1	0	0	3
We are excited to be the day 0 launch partner for Rime Mist v3! Mist v3 is a significant step forward for production v…	@baseten	Company	Original	2026-04-07	1,256	15	1	0	0	5
Our engineers just shipped the fastest named entity recognition (NER) inference on the market: 1 ms P50 and 3 ms P99 se…	@baseten	Company	Quote	2026-04-06	13,074	114	6	4	1	74
What if training LLMs didn't require rebuilding your entire product as a sandbox? RL training forces companies to repa…	@baseten	Company	Quote	2026-04-06	14,147	101	5	5	0	111
Gemma 4 is live on Baseten and available to all customers on day 0 via the Baseten model library. All models in the Ge…	@baseten	Company	Original	2026-04-02	2,467	42	7	3	0	12
What if LLMs could remember as humans do? LLM memory is either perfect and lossless or ultra-compressed. What does a s…	@baseten	Company	Quote	2026-04-01	5,704	53	9	1	0	45
We've had a great month of March! A brief recap: -> NVIDIA GTC, featuring books, ice cream, and swag -> KubeCon EMEA, …	@baseten	Company	Original	2026-04-01	922	3	0	0	1	1
Our researchers have been running autoresearch loops on Baseten Training for months. It's now a core part of how we wor…	@baseten	Company	Quote	2026-03-30	20,385	67	7	0	0	87
Initially, we believed that open-source models could only accomplish a fraction of the tasks of closed-source models. S…	@baseten	Company	Quote	2026-03-30	23,092	55	3	0	1	37
We're thrilled to be featured in @fastcompany's 2026 list of most innovative companies in applied AI. This is the year…	@baseten	Company	Original	2026-03-27	1,708	18	2	0	2	4
You've seen the Pied Piper memes, but if you want to understand how TurboQuant works you should read this article from …	@baseten	Company	Quote	2026-03-27	10,370	56	6	2	2	60
Check out GLM-5 on Baseten Model API live in @Linkup_platform, making agents both fast and accurate:	@baseten	Company	Quote	2026-03-25	1,945	14	2	2	0	1
Zed built its editor from scratch because performance is non-negotiable for a responsive IDE. When your product lives o…	@baseten	Company	Original	2026-03-24	19,543	57	4	2	3	9
Open-source LLM training is fragmented, and there's no clear guide. @stefanopopoulos on our Post-Training team mapped…	@baseten	Company	Quote	2026-03-24	9,542	67	8	3	3	59
We are thrilled to welcome Sameer Paranjpye to lead our engineering organization. Welcome, Sameer! https://t.co/ysqoU…	@baseten	Company	Original	2026-03-23	8,538	40	6	2	2	5
Cold starts for large models are one of the hardest problems in AI inference infrastructure. Today we're launching the …	@baseten	Company	Quote	2026-03-19	2,648	33	2	2	0	9
Developers used to argue about programming languages; now they argue about harnesses. NemoClaw is NVIDIA's answer to y…	@baseten	Company	Quote	2026-03-19	2,043	12	3	0	0	5
Want to enjoy a sweet treat in between sessions and meetings? Come by for a scoop of our ice cream, created exclusivel…	@baseten	Company	Original	2026-03-18	969	9	3	1	0	2
If you've signed up for a physical copy of Inference Engineering and are at GTC, come to our booth for a free copy fres…	@baseten	Company	Original	2026-03-18	1,701	14	0	1	1	2
Travel in style. Get rides to and from GTC in our cars. 👀 https://t.co/x4sMwXCw9S	@baseten	Company	Original	2026-03-18	1,814	14	2	1	1	4
Don't miss @philipkiely's talk on high-performance inference for frontier AI models at 4 PM today! https://t.co/3YkHXJk…	@baseten	Company	Original	2026-03-16	1,115	14	3	1	0	4
Live from Jensen's keynote remarks at GTC: "The inflection point of inference has arrived. AI now has to think. In or…	@baseten	Company	Original	2026-03-16	1,924	19	4	3	0	5
Happy GTC week! Need to hail a ride to or from the conference? Reply or repost with a picture of our Baseten-wrapped r…	@baseten	Company	Original	2026-03-16	2,480	35	3	1	2	1
In this piece, @maxkirkby and @oneill_c use constitutional alignment as a testbed to evaluate the importance of on- ver…	@baseten	Company	Quote	2026-03-13	970	13	0	0	0	1
We're excited for NVIDIA GTC next week! Here's an overview of what to expect: -> @philipkiely's speaking session on h…	@baseten	Company	Original	2026-03-13	2,900	16	0	0	1	2
We love seeing our customer and partner friends IRL, and we've got a lot more events coming up! https://t.co/bnv5j6FJXc…	@baseten	Company	Original	2026-03-12	734	14	1	1	0	0
We are thrilled to welcome Matt Slagle to lead our global revenue organization! Welcome, Matt. https://t.co/32eBpGME5…	@baseten	Company	Original	2026-03-12	4,068	22	2	1	1	0
We're excited to be day-0 launch partners for NVIDIA Nemotron 3 Super! You can try it now on Baseten, or read @rachelr…	@baseten	Company	Original	2026-03-11	4,933	64	7	2	2	9
The Posit team is behind some of the most widely used data science tools, including RStudio and Positron. Posit uses Ba…	@baseten	Company	Original	2026-03-10	1,447	14	2	0	0	2
you're probably worried about the wrong 9's	@baseten	Company	Quote	2026-03-09	2,296	24	0	1	0	6
We've launched the fastest GLM 5 API available at 190 TPS and 0.79 sec TTFT with the Baseten Inference Stack. Ready fo…	@baseten	Company	Original	2026-03-06	19,658	101	8	16	3	26
Long-running agents accumulate context while model memory stays fixed. This leads to a tradeoff: either discard older i…	@baseten	Company	Original	2026-03-05	2,496	39	5	6	2	15
Earlier this month, we hosted our biannual company-wide offsite and gathered 180 teammates in Austin, TX. Highlights in…	@baseten	Company	Original	2026-03-04	5,190	44	1	6	3	10
A few months ago, we trained a specialist model that beats Gemini on emergency medicine documentation and runs 6–8x fas…	@baseten	Company	Original	2026-03-03	1,209	21	2	1	0	6
We painted San Francisco green and pink, and the message is clear — you need to own your inference. If you spot us ar…	@baseten	Company	Original	2026-03-02	3,345	44	9	11	5	13
What is a core advantage of open-source? The community and economics. " If you need large manpower for something you'r…	@baseten	Company	Original	2026-02-27	747	13	0	0	0	3
Large world models (LWMs) are the latest modality in AI. These foundation models push the boundaries of what AI can do …	@baseten	Company	Quote	2026-02-25	1,609	17	0	0	0	6
Introducing RadixMLP: intra-batch prefix deduplication for 1.4–5x faster prefill. Tokens with identical prefixes (like…	@baseten	Company	Original	2026-02-24	2,951	27	3	0	2	18
New in the Baseten ML Cookbook: a multi-node training recipe for @MiniMax_AI M2.5, the open MoE model that's SOTA in co…	@baseten	Company	Original	2026-02-24	943	13	0	0	0	7
we've learned from the best!	@baseten	Company	Quote	2026-02-24	887	19	0	1	0	1
Every developer has the opportunity to own their intelligence by learning how to do inference, but today the field is s…	@baseten	Company	Quote	2026-02-23	2,082	29	1	2	0	4
"Medicine doesn’t tolerate “remarkably capable.” It requires specifically correct." We're honored to partner with Open…	@baseten	Company	Quote	2026-02-23	795	7	0	1	0	3
"No other product lets you launch ten different training jobs on four different datasets." –Head of Clinical NLP, OpenE…	@baseten	Company	Original	2026-02-20	16,895	30	7	0	2	24
GLM 5 is live on Baseten. Opus 4.6 level performance at 10% of the cost. It takes the intelligence from its predecess…	@baseten	Company	Original	2026-02-19	2,034	34	1	2	0	6
Generational AI companies are powered by Baseten. Why? We obsess over the milliseconds, so they can ship the future. …	@baseten	Company	Original	2026-02-19	4,098	37	7	4	4	11
coming soon 👀	@baseten	Company	Quote	2026-02-18	2,277	30	1	0	0	7
The Gamma team uses the Baseten Inference Stack to: -> Generate millions of images per day -> Reduce latency per image…	@baseten	Company	Original	2026-02-17	756	16	1	0	0	4
Just because it's a federal holiday doesn't mean we're slacking. MiniMax M2.5 is live on our Model APIs. Try it here:…	@baseten	Company	Original	2026-02-16	5,995	44	3	1	0	3
RL often throws away useful signal at intermediate steps, or as @karpathy put it, it's like "sucking supervision throug…	@baseten	Company	Original	2026-02-13	14,582	34	3	0	1	12
We replicated Microsoft Research's Generative Adversarial Distillation (GAD) to distill Qwen3-4B from GPT-5.2. Standar…	@baseten	Company	Original	2026-02-13	4,089	30	4	1	3	10
Sully.ai is transforming healthcare efficiency with Baseten’s Model APIs, running frontier open models like gpt-oss-12…	@baseten	Company	Original	2026-02-12	1,324	24	4	1	0	5
Following up on yesterday's release 🚨 How did we build the fastest Kimi K2.5 inference? • Custom EAGLE-3 speculator…	@baseten	Company	Quote	2026-02-11	16,860	47	7	1	2	5
Continuing this week with a case study ☕️ How did @sullyai return 30M+ clinical minutes to doctors? By ditching closed…	@baseten	Company	Original	2026-02-10	4,219	30	3	2	4	10
Introducing Kimi K2.5 on Baseten’s Model APIs with the most performant TTFT (0.26 sec) and TPS (340) on Artificial Anal…	@baseten	Company	Original	2026-02-10	15,323	98	8	11	5	21
We're living in the era of metric obsession. (How is your sleep score after Super Bowl weekend?) 🏈 Now your metric obs…	@baseten	Company	Original	2026-02-09	1,081	28	0	2	0	4
LLMs are amnesiacs. Once context fills up, they forget everything. To fight this means grappling with a core question: …	@baseten	Company	Original	2026-02-06	7,722	79	11	8	1	29
What’s the connection between LLM fine-tuning and Hollywood? It turns out that there are many, from VFX tooling to bran…	@baseten	Company	Original	2026-02-05	797	16	4	2	0	0
LLMs display human-like behavior, with Karpathy once describing them as stochastic "people spirits." This makes them n…	@baseten	Company	Original	2026-02-05	702	13	0	0	0	3
The best OpenClaw🦞 setup is fully open-source. Kimi K2.5 on Baseten outperforms Opus 4.5 on agentic benchmarks at 8x …	@baseten	Company	Original	2026-02-04	7,724	55	6	2	2	34
MARS-Flash is now available on Baseten. If you know Baseten, you know we’re obsessed with speed. Enter, MARS-Flash. MA…	@baseten	Company	Original	2026-02-04	690	12	0	1	0	0
Ready to cook? 🍳 New training recipe in the Baseten ML Cookbook: GLM 4.7 and 4.7 Flash Fine tune the leading multimod…	@baseten	Company	Original	2026-02-03	889	10	0	0	1	6
Thanks @NVIDIAAI for inviting us to Dynamo Day! We're active users of Dynamo, iterating on it in production for perform…	@baseten	Company	Original	2026-02-03	3,952	16	1	0	1	2
Nemotron 3 Nano NVFP4 is now available on Baseten + NVIDIA B200 BF16-level accuracy, up to 4× higher throughput vs FP8…	@baseten	Company	Original	2026-01-28	2,362	16	2	0	1	2
Thank you @BloombergTV for having our CEO and co-founder @tuhinone and day 1 investor @saranormous yesterday to discuss…	@baseten	Company	Original	2026-01-27	2,625	29	3	2	0	4
We boosted acceptance rate by up to 40% with the Baseten Speculation Engine. How? By combining Multi-Token Prediction …	@baseten	Company	Original	2026-01-27	13,959	29	5	1	2	6
LIVE Tune in to hear @tuhinone discuss our series E, open source, and the multi-model future on CNBC	@baseten	Company	Quote	2026-01-26	1,396	6	1	1	1	0
Last week, we announced our $300M series E at a $5B valuation. This week, we’re unpacking it all (and more!), starting…	@baseten	Company	Original	2026-01-26	2,815	31	2	0	0	2
ANNOUNCEMENT: @tuhinone IS COMING ON @tbpn RIGHT NOW WE REPEAT: @tuhinone IS COMING ON TBPN RIGHT NOW	@baseten	Company	Original	2026-01-23	1,514	26	1	0	0	0
We’re thrilled to announce that we have raised $300M at a $5B valuation. The round is led by IVP and CapitalG, both dou…	@baseten	Company	Original	2026-01-23	300,560	344	24	41	24	164
Tired of waiting for video generation? Say less. We've optimized the Wan 2.2 runtime to hit: 3x faster inference on NV…	@baseten	Company	Original	2026-01-22	1,898	19	3	4	0	3
We’re thrilled to be working with @LangChain to power the fastest way to generate production-ready agents without code.…	@baseten	Company	Original	2026-01-21	5,446	23	4	4	1	8
Want to learn about how to run high performance LLM inference at scale?Our Head of DevRel @philipkiely has the perfect …	@baseten	Company	Original	2026-01-20	2,718	26	3	0	2	6
🚀 We're thrilled to introduce the fastest, most accurate, and cost-efficient Whisper-powered transcription and diarizat…	@baseten	Company	Original	2026-01-16	2,734	28	2	7	2	4
"Engineers who are product-minded and know what they want to ship, because they understand their users, now they have i…	@baseten	Company	Original	2026-01-13	1,079	10	1	0	0	1
Inference performance isn’t just about the model. It relies on the entire inference stack. In our Inference Stack whit…	@baseten	Company	Original	2026-01-09	9,097	42	5	2	2	20
"It's going to let you take the world as you see it, or let the code that you have that works today [...] and just go."…	@baseten	Company	Original	2026-01-07	1,002	5	1	0	1	2
Here’s the part people don’t like to say out loud: in most practical settings, RL is the wrong starting point. The cri…	@baseten	Company	Original	2026-01-06	1,084	11	3	0	1	1
the fastest GLM 4.7 available to try today on Baseten https://t.co/yYLe0iKxNv	@baseten	Company	Quote	2025-12-29	2,082	13	1	1	0	3
Just in time for the new year! Awesome job by our model performance team to hit the top of @ArtificialAnlys for GLM 4…	@baseten	Company	Quote	2025-12-28	1,723	20	3	1	0	1
2025 redefined what it takes to get an AI application production-ready. Here are 3 infra learnings that will help you b…	@baseten	Company	Original	2025-12-22	888	11	2	1	0	3
@thealexker from Baseten recently spoke at AI Dev x NYC by @DeepLearningAI. In this talk, he covers the rise of open-…	@baseten	Company	Original	2025-12-19	580	7	0	1	0	0
Turns out we know a lot of people that like to go fast. Congrats to Stewart Noll and Alex Choy for taking the top tw…	@baseten	Company	Original	2025-12-17	1,219	28	1	1	0	2
Baseten supports @nvidia Nemotron 3 Nano on day zero Up to 4× faster token generation, high accuracy, and predictable …	@baseten	Company	Original	2025-12-15	2,733	21	2	1	0	1
RL is a critical step towards serving production-ready AI applications with very specific needs requirements. In this g…	@baseten	Company	Original	2025-12-12	3,038	17	4	2	1	6
"We want people to own their own intelligence, and we now see a really straight shot to get there." @amiruci sits down…	@baseten	Company	Original	2025-12-11	1,768	16	3	3	1	4
Today we’re welcoming the @parsedlabs team to Baseten! With their RL and post-training expertise, Baseten is enabling …	@baseten	Company	Original	2025-12-10	22,627	47	5	2	4	10
"These video gen models are getting pretty good. The real Hollywood use case is they messed up one of the shots and the…	@baseten	Company	Original	2025-12-09	3,754	25	5	1	1	6
"Baseten's [training] solution is not point and click. Baseten's solution caters to people who are more hands on. When …	@baseten	Company	Original	2025-12-08	2,099	32	3	1	1	2
Are you wondering how to get GPT-5-level performance at the fraction of the cost? We walk you through the top 3 directi…	@baseten	Company	Original	2025-12-05	779	11	0	0	0	2
We're excited to partner with @getstream_io to help developers build fast, production-ready Vision Agents. Together, …	@baseten	Company	Quote	2025-12-05	1,379	9	2	1	0	1
If you need an adrenaline rush to wake up from your post-Thanksgiving stupor… we got you. @deepseek_ai V3.2 dropped th…	@baseten	Company	Original	2025-12-04	4,760	43	12	10	4	2
Agents that don't hallucinate? Meet APT: @ScaledCognition's Agentic Pretrained Transformer — the only frontier model fo…	@baseten	Company	Original	2025-12-04	1,449	16	2	3	2	5
Baseten is proud to support training jobs for The LLM Data Company 💪	@baseten	Company	Quote	2025-12-03	2,040	17	4	0	1	1
Voice AI should feel like talking to a human. @rimelabs gets this and their models capture nuance, tone, and personali…	@baseten	Company	Original	2025-12-02	851	24	1	3	0	4
Congrats to the team at Mistral AI on the new Apache-licensed Mistral Large 3, a foundation model with frontier-class i…	@baseten	Company	Quote	2025-12-02	952	10	0	2	0	0
Cool 🚀 Launch alert! We’ve officially opened applications for Baseten for Startups > Up to $25K credits for inferenc…	@baseten	Company	Original	2025-12-01	33,429	38	7	3	6	14
Enterprises are adopting AI way faster than anyone expected and @Get_WRITER is leading the charge. From ROI-driven tra…	@baseten	Company	Original	2025-11-21	39,666	38	6	3	3	7
Congrats to our friends at Deep Cogito on launching the most powerful US-based OSS model. It turns out LLM self play pr…	@baseten	Company	Quote	2025-11-19	1,265	10	0	0	0	0
@tuhinone sits down on the Gradient Dissent podcast by @wandb They discuss all things inference and what sets Baseten…	@baseten	Company	Original	2025-11-19	413	7	0	1	0	0
Shoutout to the incredible team at @oxen_ai! Turning datasets → deployed models like it’s light work. They build fast. …	@baseten	Company	Original	2025-11-18	1,691	20	4	4	3	2
Happy Monday 👋 We're pleased to welcome some new Baseten crew members to the team. Say hello to Tom Berger, Paulina Pev…	@baseten	Company	Original	2025-11-17	762	5	0	0	0	0
@thealexker just spoke on stage at @DeepLearningAI AI Dev 25 x NYC about how Baseten helped Sourcegraph optimize their …	@baseten	Company	Original	2025-11-14	174	8	0	0	0	0
We're here at AI Dev 25 x NYC with @DeepLearningAI and @AndrewYNg Come say hi! https://t.co/CPVi2GFsCR	@baseten	Company	Original	2025-11-14	1,649	17	2	1	1	2
Working with the @GammaApp team never quite feels like work, and that’s how their product feels. "Criminally fun." We …	@baseten	Company	Original	2025-11-13	22,867	44	4	1	3	7
Baseten used @nvidia Dynamo to double inference speed for long-context code generation and increased throughput by 1.6x…	@baseten	Company	Quote	2025-11-13	1,214	16	0	1	0	3
From cricket to the earliest days of Baseten to where we are today, @adityaag and @tuhinone go deep on the Minus One po…	@baseten	Company	Quote	2025-11-13	2,619	18	2	0	0	1
Welcome to the new age Defense Against the Dark Arts. It's called fast inference! (& Harry Potter would be jealous). …	@baseten	Company	Original	2025-11-12	1,922	26	2	4	2	4
Congrats to the World Labs team on the launch today! Marble lets you create 3D worlds from just a single image, text pr…	@baseten	Company	Quote	2025-11-12	3,070	33	2	6	4	3
We had a great time at our Baseten rooftop happy hour at KubeCon with our friends at @OpsLevelHQ, @Sentry, @JellyFish_W…	@baseten	Company	Original	2025-11-12	963	21	0	0	1	1
At @KubeCon_ ? Swing by Booth #631 to test your inference knowledge and earn some swag! https://t.co/rLCW9dvr1o	@baseten	Company	Original	2025-11-11	913	17	0	0	0	0
It’s Monday, and we could all use a little help thinking. Thankfully we have the new Kimi K2 Thinking to do it for us. …	@baseten	Company	Original	2025-11-11	60,790	97	11	7	10	25
Excited to share this piece from @VentureBeat spotlighting how Baseten is redefining the AI infrastructure game: “Bas…	@baseten	Company	Original	2025-11-10	1,350	17	7	2	0	0
Congratulations on the launch, this will be big! Excited to growth the partnership 🤝	@baseten	Company	Quote	2025-11-10	1,540	21	0	1	0	1
Congratulations to @GammaApp on this amazing milestone! We're happy to continue our support of your amazing growth. …	@baseten	Company	Quote	2025-11-10	1,850	13	2	0	0	2
Heading to KubeCon next week? Come visit the team at Booth #631 to test your AI knowledge. Top of the leaderboard gets …	@baseten	Company	Original	2025-11-07	673	9	0	0	0	1
Fun fact - we asked people to describe their favorite agent in SF. We got suggestions for a bunch of new agentic apps t…	@baseten	Company	Original	2025-11-05	653	13	0	1	0	1
@Madisonkanna sits down with @thdxr from @opencode to talk about agents, code gen, and open source. Check out the full…	@baseten	Company	Original	2025-11-05	12,435	50	4	1	0	14
Our team grows when our customers grow, and our customers are on a tear. Please welcome Zane, Allen, Brooke, and Katie …	@baseten	Company	Original	2025-11-04	835	13	0	0	0	2
After months of feedback from our early customers and thousands of jobs completed, Baseten Training is officially ready…	@baseten	Company	Original	2025-10-30	14,573	42	5	4	8	8
We had a great weekend at @CalHacks 12.0 in SF Congratulations to the winners of the Baseten prize for best use of ope…	@baseten	Company	Original	2025-10-29	698	10	1	0	0	1
We are so excited to be a launch partner for @nvidia Nemotron Nano 2 VL today and offer day-zero support for this high…	@baseten	Company	Original	2025-10-28	1,900	21	3	2	0	1
Another week, another group of new Baseten colleagues to introduce! Please welcome Victoria Jones, Analisa Ruff, Aryan …	@baseten	Company	Original	2025-10-27	1,401	20	0	2	0	1

Plushcap, by Matt Makai. 2021-2026.