| RT @tensorfiend: @vast_ai GPU renting prices are very low and reasonable. Perfect for my experiments👌. |
@vast_ai |
Company |
Repost |
2026-10-29 |
99 |
0 |
1 |
0 |
0 |
0 |
| Runpod's GraphQL API is rate-limited starting today. REST v1 retires Nov 15.
Updating your workflows? Explore 68+ GPU … |
@vast_ai |
Company |
Original |
2026-10-27 |
613 |
3 |
1 |
0 |
0 |
0 |
| RT @varien: "seedance is so expensive oh no"
my entire movie studio runs locally. comfyui, open weights, loras, video,… |
@vast_ai |
Company |
Repost |
2026-10-26 |
131 |
0 |
30 |
0 |
0 |
0 |
| https://t.co/fWRuSQeT0Y |
@vast_ai |
Company |
Quote |
2026-10-26 |
726 |
9 |
0 |
2 |
0 |
0 |
| Proud to be among @brexHQ’s 25 fastest-growing software vendors this fall. Thanks to the builders choosing Vast and the… |
@vast_ai |
Company |
Quote |
2026-10-21 |
761 |
8 |
3 |
1 |
0 |
0 |
| 650,000 compute transactions ran through Vast in the last 30 days. Today we filed a comment with the @CFTC on compute d… |
@vast_ai |
Company |
Original |
2026-10-21 |
819 |
4 |
3 |
1 |
1 |
0 |
| RT @Natemarmorstein: Check us out @vast_ai, we currently have 40 B200s from $6.25/hr, 20+ H100s from $1.79/hr, and 100+… |
@vast_ai |
Company |
Repost |
2026-10-20 |
105 |
0 |
1 |
0 |
0 |
0 |
| okay, but how much VRAM? |
@vast_ai |
Company |
Quote |
2026-10-16 |
1,499 |
24 |
1 |
0 |
0 |
0 |
| Our COO Travis Cannell joined @WCMC_Podcast to talk about how the GPU market works.
https://t.co/Vxk8tVBaIH |
@vast_ai |
Company |
Original |
2026-10-15 |
907 |
4 |
0 |
2 |
0 |
0 |
| https://t.co/6FdYZbgYJe |
@vast_ai |
Company |
Original |
2026-10-15 |
553 |
0 |
0 |
1 |
0 |
1 |
| RT @Natemarmorstein: Excited to share that I've joined @vast_ai as a Developer Relations Engineer!
Looking forward to… |
@vast_ai |
Company |
Repost |
2026-10-15 |
117 |
0 |
1 |
0 |
0 |
0 |
| RT @vast_ai: We’re hosting LA Tech Week.
Thursday 10/15, 6-8pm at our Westwood office.
Food, open bar, and a conversa… |
@vast_ai |
Company |
Repost |
2026-10-14 |
-- |
0 |
3 |
0 |
0 |
0 |
| RT @vast_ai: We’re hosting LA Tech Week.
Thursday 10/15, 6-8pm at our Westwood office.
Food, open bar, and a conversa… |
@vast_ai |
Company |
Repost |
2026-10-13 |
-- |
0 |
3 |
0 |
0 |
0 |
| We’re hosting LA Tech Week.
Thursday 10/15, 6-8pm at our Westwood office.
Food, open bar, and a conversation about wh… |
@vast_ai |
Company |
Original |
2026-10-12 |
748 |
6 |
3 |
3 |
0 |
1 |
| 84 GB is enough to hold some 27B-class MoE models at full context that would otherwise need two RTX 5090s. That’s the c… |
@vast_ai |
Company |
Original |
2026-10-10 |
1,045 |
23 |
4 |
5 |
1 |
1 |
| New MiniMax M3 template now available in the https://t.co/DfwOtzr83c model library
@MiniMax_AI M3 is a native multimod… |
@vast_ai |
Company |
Original |
2026-10-06 |
1,297 |
16 |
3 |
2 |
0 |
2 |
| Come find us at The AI Conference in San Francisco @ Pier 48. Booth 202. Last day!
We've got a Nintendo Switch 2 raffl… |
@vast_ai |
Company |
Original |
2026-10-01 |
1,242 |
29 |
3 |
12 |
0 |
0 |
| Setting up an AI agent and not sure where to start? Here's a walkthrough on picking a framework and model, sizing your … |
@vast_ai |
Company |
Original |
2026-09-29 |
731 |
4 |
1 |
5 |
0 |
0 |
| https://t.co/DfwOtzr83c is at The AI Conference 2026.
September 29 to October 1. Pier 48, San Francisco. Booth 202. … |
@vast_ai |
Company |
Original |
2026-09-21 |
956 |
8 |
1 |
1 |
0 |
1 |
| Renting GPU compute beats buying when your LLM workload is bursty or changing.
No hardware spend. Pick the GPU for the… |
@vast_ai |
Company |
Original |
2026-09-15 |
892 |
17 |
1 |
2 |
0 |
1 |
| RT @traviscannell: @vast_ai is over 90% utilized! Amazing seeing the fleet cooking.
Bring your GPUs asap to realize t… |
@vast_ai |
Company |
Repost |
2026-09-09 |
134 |
0 |
1 |
0 |
0 |
0 |
| Nvidia reportedly agreed to buy Hugging Face for $12.9B. Last funding round: $4.5B valuation. They turned down Nvidia's… |
@vast_ai |
Company |
Original |
2026-08-29 |
768 |
6 |
0 |
1 |
0 |
0 |
| New Muse Glimmer 30B template available in the model library
Muse Glimmer 30B by @Meta is a multimodal reasoning model… |
@vast_ai |
Company |
Original |
2026-08-14 |
1,837 |
21 |
3 |
5 |
0 |
0 |
| MiniMax H3 Now Available in the Model Library
@MiniMax_AI H3 is an omni-modal generative system from MiniMaxAI that pr… |
@vast_ai |
Company |
Original |
2026-08-11 |
2,010 |
15 |
3 |
3 |
0 |
2 |
| DeepSeek V4 Flash Now Available in the Model Library
DeepSeek V4 Flash is a Mixture-of-Experts (MoE) language model wi… |
@vast_ai |
Company |
Original |
2026-08-10 |
2,236 |
17 |
3 |
8 |
0 |
1 |
| New on Vast: hundreds of B200s live now with more coming through September, plus Kimi K3 (the first open-weight 3T-clas… |
@vast_ai |
Company |
Original |
2026-08-07 |
2,132 |
25 |
4 |
5 |
0 |
2 |
| Hugging Face Storage Buckets are now on https://t.co/DfwOtzr83c
Connect your HF Storage Bucket as a Cloud Connection i… |
@vast_ai |
Company |
Original |
2026-08-06 |
32,338 |
84 |
26 |
10 |
3 |
12 |
| NVIDIA's DGX Spark packs a Grace Blackwell GPU and 128GB unified memory into a box the size of a Mac Mini. Early review… |
@vast_ai |
Company |
Original |
2026-08-05 |
2,026 |
15 |
3 |
3 |
0 |
1 |
| Kimi K3 template now available in the https://t.co/DfwOtzr83c model library
Kimi K3 is an open-weight, native multimod… |
@vast_ai |
Company |
Original |
2026-07-31 |
4,149 |
32 |
3 |
13 |
1 |
5 |
| NVIDIA DGX Spark: 6"x6"x2", 128GB unified memory, models up to 200B parameters. A supercomputer that fits in your hand.… |
@vast_ai |
Company |
Original |
2026-07-28 |
2,679 |
10 |
1 |
4 |
0 |
0 |
| How much does a GPU actually cost to rent right now?
We put together a live pricing guide comparing real-time rates ac… |
@vast_ai |
Company |
Original |
2026-07-23 |
3,849 |
13 |
0 |
6 |
1 |
1 |
| Token-based API pricing charges you per word your agent thinks. Rent GPUs and self-host instead, and you pay for comput… |
@vast_ai |
Company |
Original |
2026-07-09 |
5,222 |
24 |
3 |
7 |
0 |
1 |
| https://t.co/2WHvjspuyh is hiring.
We're growing. Fast.
Agent-ready AI infrastructure. Profitable, venture-backed. Ne… |
@vast_ai |
Company |
Original |
2026-07-06 |
23,790 |
79 |
8 |
27 |
0 |
17 |
| NVIDIA Rubin is official: 6 chips, rack-scale design, datacenter-as-one-computer.
50 petaFLOPS NVFP4. 260 TB/s per-rack… |
@vast_ai |
Company |
Original |
2026-06-30 |
7,270 |
28 |
5 |
7 |
0 |
3 |
| One embedding model. Multiple vector sizes. No retraining.
Matryoshka embeddings let you tune retrieval cost vs. quali… |
@vast_ai |
Company |
Original |
2026-06-27 |
5,780 |
18 |
2 |
3 |
0 |
1 |
| You can now rent GPUs from inside your coding agent.
Vast ships as a plugin for Claude Code, Codex, and Cursor. Search… |
@vast_ai |
Company |
Original |
2026-06-24 |
5,723 |
21 |
4 |
7 |
1 |
2 |
| running large model inference on h200 80GB HBM3e for $2.89/hr if you catch the p10. Other clouds averaging $8.46 for th… |
@vast_ai |
Company |
Original |
2026-06-24 |
4,088 |
13 |
0 |
2 |
0 |
5 |
| people keep saying h100s are holding steady but they dropped 8% this week. $2.15 median, $1.87 at p10. Supply is tight … |
@vast_ai |
Company |
Original |
2026-06-23 |
3,676 |
8 |
0 |
2 |
0 |
1 |
| rtx 5090 at $0.50/hr median with 1,244 available. Prices down 6% from last week. |
@vast_ai |
Company |
Original |
2026-06-19 |
4,671 |
11 |
0 |
4 |
0 |
2 |
| deepseek v3 on 8xH200, $3.42/hr. 211 instances available right now across the fleet |
@vast_ai |
Company |
Original |
2026-06-18 |
4,176 |
11 |
0 |
1 |
0 |
0 |
| We built a fully functional coding agent in under 100 lines of Python. No framework. No magic. Just the model requestin… |
@vast_ai |
Company |
Original |
2026-06-17 |
3,682 |
3 |
0 |
1 |
0 |
2 |
| A100 80GB median is $1.05/hr, bottom decile hits $0.67/hr. 141 cards live. Solid option if you're doing multi-GPU train… |
@vast_ai |
Company |
Original |
2026-06-17 |
2,929 |
6 |
0 |
0 |
0 |
1 |
| NEW Installer Wizard for Hosts
This TUI offers a new guided terminal setup that replaces the CLI script. It walks you … |
@vast_ai |
Company |
Original |
2026-06-16 |
2,834 |
13 |
2 |
2 |
0 |
3 |
| B300 has 2x the memory of H200 (288 GB vs 141 GB) and 8 TB/s bandwidth. But is it actually worth the 1,400W TGP for you… |
@vast_ai |
Company |
Original |
2026-06-16 |
2,226 |
6 |
0 |
2 |
0 |
0 |
| running flux.1 today, 12B fits clean on a single 4090. $0.44/hr with 1475 units up across the fleet. |
@vast_ai |
Company |
Original |
2026-06-13 |
2,805 |
6 |
1 |
1 |
0 |
0 |
| June Product Update
NVIDIA B300 GPUs are now available on https://t.co/DfwOtzqAdE, alongside hundreds of new B200s.
T… |
@vast_ai |
Company |
Original |
2026-06-11 |
3,113 |
11 |
1 |
2 |
0 |
0 |
| B200s dropped 27% in a week, now $4.38/hr median. 252 available. New supply is landing fast on these. |
@vast_ai |
Company |
Original |
2026-06-11 |
2,121 |
3 |
0 |
0 |
0 |
1 |
| The NVIDIA B300 has 288 GB of HBM3e memory. That's 50% more than the B200, 2x the H200, and 3.6x the H100.
A single GP… |
@vast_ai |
Company |
Original |
2026-06-06 |
3,831 |
12 |
0 |
7 |
0 |
0 |
| 5090s at $0.67/hr if you catch the p10. 32GB GDDR7 is enough for most LoRA jobs without offloading anything |
@vast_ai |
Company |
Original |
2026-06-06 |
2,230 |
11 |
0 |
0 |
0 |
0 |
| ```python
client.instance.create(gpu_type="H200", num_gpus=8, image="vllm/vllm-openai", env={"MODEL": "moonshotai/Kimi-… |
@vast_ai |
Company |
Original |
2026-06-05 |
2,326 |
7 |
2 |
0 |
0 |
2 |
| Great to see Vast ai back on Ramp’s Top Software Vendors list alongside some of the fastest-growing names in AI and sof… |
@vast_ai |
Company |
Quote |
2026-06-04 |
1,661 |
5 |
0 |
0 |
0 |
2 |
| What actually is a neocloud?
AI-first GPU cloud built for ML workloads — not bolted on after the fact. The business mo… |
@vast_ai |
Company |
Original |
2026-06-03 |
1,412 |
5 |
1 |
4 |
0 |
0 |
| B200 median jumped from $4.17 to $7.23 in a week. 91% utilization across the fleet, 14 cards still open. |
@vast_ai |
Company |
Original |
2026-05-30 |
2,315 |
6 |
0 |
1 |
0 |
0 |
| h100 80GB at $2.00/hr if you catch the p10. Running a full 70B LoRA training loop on 2 cards, no quantization needed. 3… |
@vast_ai |
Company |
Original |
2026-05-29 |
2,688 |
7 |
0 |
2 |
0 |
0 |
| Now Available on https://t.co/DfwOtzr83c: NVIDIA B300
Blackwell Ultra architecture, 288GB HBM3e memory, 20,480 CUDA co… |
@vast_ai |
Company |
Original |
2026-05-28 |
2,604 |
12 |
0 |
0 |
0 |
2 |
| The RTX 5090 has been out for over a year and you still can't buy one at MSRP.
We wrote the full spec breakdown anyway… |
@vast_ai |
Company |
Original |
2026-05-28 |
1,663 |
10 |
0 |
3 |
0 |
0 |
| deepseek r1 full 685B on 8xH200, no quantization needed. 250 instances up right now at $3.95/hr |
@vast_ai |
Company |
Original |
2026-05-26 |
2,096 |
6 |
0 |
3 |
0 |
0 |
| running flux1 on a 4090 costs $0.40/hr median. 48 cards up right now, 24GB is plenty for most diffusion workloads. |
@vast_ai |
Company |
Original |
2026-05-21 |
3,357 |
8 |
1 |
4 |
0 |
3 |
| Unsloth Studio is available on Vast.
500+ models. 2x faster training. 70% less VRAM. No accuracy loss. Fine-tune witho… |
@vast_ai |
Company |
Original |
2026-05-20 |
3,084 |
15 |
0 |
5 |
0 |
2 |
| ```python
client.launch(
model="deepseek-ai/DeepSeek-V3-0324",
num_gpus=8,
gpu_type="H200"
)
```
DeepSeek V3 fill… |
@vast_ai |
Company |
Original |
2026-05-20 |
2,231 |
9 |
0 |
5 |
0 |
0 |
| Your GPU is sitting idle. 120,000+ developers are looking for exactly what you have.
RTX 5090 rig (4x GPUs, 80% utiliz… |
@vast_ai |
Company |
Original |
2026-05-20 |
2,250 |
10 |
1 |
2 |
0 |
6 |
| h100 at $2.51/hr median, up 18% from last week. 3 available across the whole fleet. If you locked in at $1.53 p10, nice… |
@vast_ai |
Company |
Original |
2026-05-16 |
3,678 |
14 |
1 |
7 |
0 |
4 |
| 8 creative AI tools. One container. Zero setup headaches.
The All-in-One App Studio gives you ComfyUI, SD Forge, Wan2G… |
@vast_ai |
Company |
Original |
2026-05-15 |
4,260 |
10 |
1 |
2 |
0 |
1 |
| people keep saying h200s are $8+/hr. We have them at $3.87, p10 is $3.35. 2 available right now, 80GB HBM3e each. Good … |
@vast_ai |
Company |
Original |
2026-05-14 |
3,279 |
11 |
0 |
1 |
0 |
3 |
| H100s at 95% utilization with just 12 available on the platform. B200 prices falling (down 4% WoW) while consumer-tier … |
@vast_ai |
Company |
Original |
2026-05-13 |
2,298 |
7 |
1 |
5 |
0 |
0 |
| FLUX.1 (12B) on a single RTX 4090 — 24GB VRAM is enough, $0.34/hr, 403 available right now. Apache 2.0 so you can ship … |
@vast_ai |
Company |
Original |
2026-05-08 |
4,102 |
13 |
2 |
9 |
0 |
1 |
| ⚡️ https://t.co/DfwOtzr83c is back on Ramp's Top SaaS Vendors trending list for May — #6, up from #9 in March.
Ramp tr… |
@vast_ai |
Company |
Original |
2026-05-06 |
3,949 |
21 |
9 |
1 |
3 |
6 |
| This week on Vast: H100 utilization hit 99%. Two GPUs left. Price climbed 14% to $2.13/hr. RTX 3090s moved the other wa… |
@vast_ai |
Company |
Original |
2026-05-06 |
2,575 |
15 |
0 |
5 |
0 |
2 |
| The KV cache is the silent VRAM killer in long-context inference. TurboQuant attacks it directly: up to 6x compression,… |
@vast_ai |
Company |
Original |
2026-05-05 |
2,124 |
11 |
0 |
4 |
0 |
1 |
| Kimi K2: 1T param MoE with 32B active params, MIT licensed. Strong on code and reasoning. Fits on 8xH200 with room to s… |
@vast_ai |
Company |
Original |
2026-05-01 |
3,526 |
18 |
1 |
7 |
0 |
1 |
| We've shipped a major upgrade to two-factor authentication on Vast.
Five things to know:
→ Authenticator app support… |
@vast_ai |
Company |
Original |
2026-04-30 |
3,588 |
19 |
1 |
7 |
0 |
2 |
| #ICYMI Mistral Small 4 unifies instruct, reasoning, and coding into a single set of weights — and you can deploy it tod… |
@vast_ai |
Company |
Original |
2026-04-29 |
2,297 |
8 |
0 |
2 |
0 |
3 |
| Deploy Llama, DeepSeek, or your own model — start for $5.
https://t.co/lkXU6qK3kE |
@vast_ai |
Company |
Original |
2026-04-28 |
2,192 |
6 |
0 |
5 |
0 |
0 |
| A100 80GB HBM2e — still the workhorse for multi-GPU fine-tuning. 37 available right now from $0.61/hr, median $1.06/hr.… |
@vast_ai |
Company |
Original |
2026-04-25 |
3,250 |
14 |
1 |
3 |
1 |
0 |
| KIMI K2.6 Template now available
Kimi K2.6 is an open-source, native multimodal agentic model from @Kimi_Moonshot tha… |
@vast_ai |
Company |
Original |
2026-04-24 |
2,764 |
17 |
2 |
2 |
0 |
3 |
| Search. Provision. Scale. Three lines of Python.
pip install vastai
from vastai_sdk import VastAI
vast = VastAI(api_k… |
@vast_ai |
Company |
Original |
2026-04-24 |
39,775 |
71 |
4 |
18 |
2 |
15 |
| DeepSeek R1 (685B) fits on 8xH200 with room to spare — 141GB per GPU across 8 cards. MIT licensed, strong on math and c… |
@vast_ai |
Company |
Original |
2026-04-23 |
4,137 |
15 |
0 |
4 |
0 |
3 |
| Gemma 4 Templates now available
Gemma 4 is @GoogleDeepMind's next-generation family of open multimodal models.
The 26… |
@vast_ai |
Company |
Original |
2026-04-21 |
4,964 |
22 |
1 |
10 |
0 |
2 |
| Weekly GPU pulse: H100 and B200 both up ~7% this week — demand climbing as teams scale training runs. Meanwhile RTX 509… |
@vast_ai |
Company |
Original |
2026-04-21 |
2,583 |
8 |
0 |
7 |
0 |
0 |
| RTX 5090 — 32GB GDDR7, starting at $0.32/hr. Fits a 13B model in FP16 with room for a 4k context window, or runs SDXL w… |
@vast_ai |
Company |
Original |
2026-04-18 |
6,127 |
21 |
2 |
9 |
0 |
7 |
| NEW All-in-One Application Studio by https://t.co/DfwOtzr83c
This template gives you a complete creative AI studio wit… |
@vast_ai |
Company |
Original |
2026-04-17 |
349,043 |
26 |
7 |
8 |
0 |
14 |
| Running DeepSeek V3 at home isn't happening — 685B MoE needs ~640GB VRAM minimum. 8xH200 (1.1TB HBM3e) handles it clean… |
@vast_ai |
Company |
Original |
2026-04-16 |
5,204 |
14 |
0 |
3 |
0 |
6 |
| H200 prices crept up 5% this week to $2.71/hr median as demand stays steady. Budget play: A100s at $0.61/hr (p10) remai… |
@vast_ai |
Company |
Original |
2026-04-14 |
6,234 |
15 |
1 |
3 |
0 |
4 |
| H200 on Vast: $2.58/hr
Same GPU on major clouds: $6.31+/hr
59% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-04-11 |
7,209 |
29 |
0 |
18 |
0 |
4 |
| 2,796+ GPUs available right now on Vast
Highest availability:
RTX 5090: 1,248 GPUs
RTX 4090: 881 GPUs
RTX 3090: 230 GP… |
@vast_ai |
Company |
Original |
2026-04-11 |
6,239 |
19 |
0 |
10 |
0 |
2 |
| H100 80GB HBM3 on Vast right now: $1.67/hr typical, $1.41/hr if you're flexible on location. 65 available. Sweet spot f… |
@vast_ai |
Company |
Original |
2026-04-11 |
4,782 |
19 |
5 |
0 |
0 |
2 |
| Only 10 L40S GPUs left on Vast at $0.56/hr.
48GB GDDR6X, ideal for inference serving.
Prices rise as supply tightens. |
@vast_ai |
Company |
Original |
2026-04-10 |
2,944 |
4 |
0 |
0 |
0 |
1 |
| We feel @1a1n1d1y 's pain. As of today, vastai destroy instance now requires confirmation before deleting. Your H100 bu… |
@vast_ai |
Company |
Quote |
2026-04-10 |
3,110 |
8 |
3 |
0 |
0 |
0 |
| NEW @UnslothAI Studio Template Now Available
Unsloth Studio is an open-source, no-code web UI for running and training… |
@vast_ai |
Company |
Original |
2026-04-10 |
3,138 |
21 |
3 |
1 |
0 |
6 |
| GPU Market — Apr 10
H200: $2.71/hr (80 avail)
H100: $1.67/hr (61 avail)
B200: $4.06/hr (167 avail)
A100: $1.06/hr (49 … |
@vast_ai |
Company |
Original |
2026-04-10 |
1,536 |
7 |
0 |
1 |
0 |
0 |
| H200 on Vast: $2.84/hr
Same GPU on major clouds: $6.31+/hr
55% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-04-09 |
1,226 |
4 |
0 |
0 |
0 |
0 |
| 2,182+ GPUs available right now on Vast
Highest availability:
RTX 5090: 850 GPUs
RTX 4090: 700 GPUs
RTX 3090: 344 GPUs… |
@vast_ai |
Company |
Original |
2026-04-09 |
1,241 |
3 |
0 |
0 |
0 |
0 |
| FLUX.1 (12B) is probably the best open image model right now for prompt adherence. Apache 2.0 licensed, fits comfortabl… |
@vast_ai |
Company |
Original |
2026-04-09 |
2,303 |
22 |
0 |
4 |
0 |
8 |
| Only 28 L40S GPUs left on Vast at $0.53/hr.
48GB GDDR6X, ideal for inference serving.
Prices rise as supply tightens. |
@vast_ai |
Company |
Original |
2026-04-08 |
1,132 |
6 |
1 |
0 |
0 |
1 |
| GPU Market — Apr 08
H200: $2.58/hr (13 avail)
H100: $1.67/hr (41 avail)
B200: $4.06/hr (88 avail)
A100: $1.05/hr (29 a… |
@vast_ai |
Company |
Original |
2026-04-08 |
1,172 |
8 |
1 |
0 |
0 |
0 |
| H200 on Vast: $2.32/hr
Same GPU on major clouds: $6.31+/hr
63% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-04-07 |
924 |
6 |
0 |
0 |
0 |
2 |
| 1,578+ GPUs available right now on Vast
Highest availability:
RTX 5090: 764 GPUs
RTX 4090: 427 GPUs
RTX 3090: 233 GPUs… |
@vast_ai |
Company |
Original |
2026-04-07 |
918 |
4 |
1 |
0 |
0 |
2 |
| H200 supply is at 97% utilization with only 14 available, yet prices dropped 11% to $2.32/hr. B200s down 24% to $3.81/h… |
@vast_ai |
Company |
Original |
2026-04-07 |
906 |
6 |
0 |
0 |
0 |
0 |
| Only 13 H200 GPUs left on Vast at $2.32/hr.
80GB HBM3e, ideal for llm training.
Prices rise as supply tightens. |
@vast_ai |
Company |
Original |
2026-04-06 |
878 |
5 |
0 |
0 |
0 |
0 |
| GPU Market — Apr 06
H200: $2.32/hr (36 avail)
H100: $1.67/hr (39 avail)
B200: $3.81/hr (122 avail)
A100: $1.05/hr (7 a… |
@vast_ai |
Company |
Original |
2026-04-06 |
825 |
6 |
0 |
2 |
0 |
2 |
| H200 on Vast: $2.32/hr
Same GPU on major clouds: $6.31+/hr
63% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-04-05 |
982 |
11 |
0 |
0 |
0 |
3 |
| 1,915+ GPUs available right now on Vast
Highest availability:
RTX 5090: 997 GPUs
RTX 4090: 443 GPUs
RTX 3090: 195 GPUs… |
@vast_ai |
Company |
Original |
2026-04-05 |
679 |
2 |
0 |
0 |
0 |
1 |
| GPU Market — Apr 04
H200: $2.32/hr (86 avail)
H100: $1.67/hr (44 avail)
B200: $3.81/hr (175 avail)
A100: $1.04/hr (20 … |
@vast_ai |
Company |
Original |
2026-04-04 |
694 |
5 |
1 |
1 |
0 |
0 |
| H200 on Vast: $2.32/hr
Same GPU on major clouds: $6.31+/hr
63% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-04-03 |
577 |
4 |
1 |
0 |
0 |
0 |
| https://t.co/2WHvjspuyh Serverless SDK is in open beta.
Decorate a function with @ remote. Call it like local Python. … |
@vast_ai |
Company |
Original |
2026-04-03 |
616 |
12 |
0 |
2 |
0 |
3 |
| 1,796+ GPUs available right now on Vast
Highest availability:
RTX 5090: 800 GPUs
RTX 4090: 441 GPUs
RTX 3090: 275 GPUs… |
@vast_ai |
Company |
Original |
2026-04-03 |
509 |
8 |
0 |
0 |
0 |
0 |
| Kimi K2: 1T param MoE with 32B active params, MIT licensed. Strong on code and reasoning benchmarks. Fits on 8xH200 (1.… |
@vast_ai |
Company |
Original |
2026-04-02 |
595 |
3 |
0 |
1 |
1 |
0 |
| H200 quick math: 141GB HBM3e fits a 70B model in BF16 with room for KV cache at full 128k context. $2.32/hr on Vast rig… |
@vast_ai |
Company |
Original |
2026-03-28 |
671 |
4 |
2 |
0 |
0 |
1 |
| Add Vast AI! |
@vast_ai |
Company |
Quote |
2026-03-26 |
708 |
2 |
0 |
0 |
0 |
0 |
| Running DeepSeek R1 at full 685B params? You need 8xH200s minimum — anything less and you're quantizing away the reason… |
@vast_ai |
Company |
Original |
2026-03-26 |
2,008 |
10 |
4 |
1 |
0 |
5 |
| RTX 3090 prices rose 7% since yesterday — now $0.15/hr on Vast.
439 available, 24GB GDDR6X. No minimum commitment. |
@vast_ai |
Company |
Original |
2026-03-23 |
4,793 |
11 |
1 |
3 |
0 |
3 |
| GPU Market — Mar 23
H200: $2.58/hr (142 avail)
H100: $1.60/hr (42 avail)
B200: $9.38/hr (135 avail)
A100: $0.99/hr (19… |
@vast_ai |
Company |
Original |
2026-03-23 |
4,202 |
14 |
4 |
4 |
0 |
9 |
| H200 on Vast: $2.58/hr
Same GPU on major clouds: $6.31+/hr
59% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-03-22 |
1,655 |
6 |
0 |
1 |
0 |
1 |
| 3,024+ GPUs available right now on Vast
Highest availability:
RTX 4090: 1,252 GPUs
RTX 5090: 817 GPUs
RTX 3090: 500 GP… |
@vast_ai |
Company |
Original |
2026-03-22 |
1,876 |
8 |
1 |
2 |
0 |
0 |
| Only 47 A100 GPUs left on Vast at $0.96/hr.
80GB HBM2e, ideal for fine-tuning.
Prices rise as supply tightens. |
@vast_ai |
Company |
Original |
2026-03-21 |
1,546 |
9 |
0 |
0 |
1 |
2 |
| GPU Market — Mar 21
H200: $2.71/hr (138 avail)
H100: $1.60/hr (56 avail)
B200: $9.38/hr (95 avail)
A100: $0.96/hr (33 … |
@vast_ai |
Company |
Original |
2026-03-21 |
1,207 |
3 |
1 |
1 |
0 |
2 |
| H200 on Vast: $2.84/hr
Same GPU on major clouds: $6.31+/hr
55% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-03-20 |
2,159 |
12 |
3 |
4 |
1 |
2 |
| We got you! https://t.co/pYWFc6wQ8q |
@vast_ai |
Company |
Quote |
2026-03-20 |
6,835 |
18 |
2 |
2 |
0 |
3 |
| GPU Market — Mar 19
H200: $2.84/hr (17 avail)
H100: $1.60/hr (12 avail)
B200: $9.38/hr (2 avail)
A100: $0.96/hr (18 av… |
@vast_ai |
Company |
Original |
2026-03-19 |
1,645 |
6 |
0 |
1 |
1 |
1 |
| H200 on Vast: $2.84/hr
Same GPU on major clouds: $6.31+/hr
55% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-03-18 |
3,088 |
7 |
3 |
0 |
0 |
1 |
| 1,984+ GPUs available right now on Vast
Highest availability:
RTX 4090: 710 GPUs
RTX 5090: 677 GPUs
RTX 3090: 429 GPUs… |
@vast_ai |
Company |
Original |
2026-03-18 |
2,409 |
8 |
1 |
2 |
0 |
1 |
| GPU Market — Mar 17
H200: $2.71/hr (41 avail)
H100: $1.60/hr (30 avail)
B200: $9.38/hr (24 avail)
A100: $0.96/hr (22 a… |
@vast_ai |
Company |
Original |
2026-03-17 |
1,776 |
6 |
1 |
2 |
0 |
2 |
| H200 on Vast: $2.45/hr
Same GPU on major clouds: $6.31+/hr
61% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-03-16 |
2,588 |
10 |
2 |
2 |
0 |
1 |
| 2,862+ GPUs available right now on Vast
Highest availability:
RTX 5090: 1,290 GPUs
RTX 4090: 887 GPUs
RTX 3090: 509 GP… |
@vast_ai |
Company |
Original |
2026-03-16 |
3,141 |
7 |
0 |
4 |
0 |
2 |
| Only 5 B200 GPUs left on Vast at $9.38/hr.
192GB HBM3e, ideal for fine-tuning.
Prices rise as supply tightens. |
@vast_ai |
Company |
Original |
2026-03-15 |
1,876 |
4 |
0 |
0 |
0 |
0 |
| GPU Market — Mar 15
H200: $2.45/hr (52 avail)
H100: $1.60/hr (40 avail)
B200: $9.38/hr (3 avail)
A100: $0.93/hr (45 av… |
@vast_ai |
Company |
Original |
2026-03-15 |
1,977 |
5 |
0 |
3 |
0 |
0 |
| 👀 |
@vast_ai |
Company |
Quote |
2026-03-15 |
2,275 |
10 |
0 |
1 |
0 |
2 |
| H200 on Vast: $2.45/hr
Same GPU on major clouds: $6.31+/hr
61% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-03-14 |
999 |
4 |
0 |
1 |
0 |
0 |
| 2,780+ GPUs available right now on Vast
Highest availability:
RTX 5090: 1,174 GPUs
RTX 4090: 780 GPUs
RTX 3090: 518 GP… |
@vast_ai |
Company |
Original |
2026-03-14 |
1,379 |
5 |
0 |
1 |
0 |
1 |
| Only 3 B200 GPUs left on Vast at $3.13/hr.
192GB HBM3e, ideal for fine-tuning.
Prices rise as supply tightens. |
@vast_ai |
Company |
Original |
2026-03-13 |
1,286 |
4 |
0 |
2 |
0 |
0 |
| GPU Market — Mar 13
H200: $2.45/hr (145 avail)
H100: $1.60/hr (24 avail)
B200: $3.13/hr (4 avail)
A100: $0.96/hr (38 a… |
@vast_ai |
Company |
Original |
2026-03-13 |
1,632 |
8 |
0 |
5 |
0 |
5 |
| H200 on Vast: $2.45/hr
Same GPU on major clouds: $6.31+/hr
61% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-03-12 |
2,053 |
20 |
1 |
6 |
0 |
1 |
| 2,219+ GPUs available right now on Vast
Highest availability:
RTX 4090: 809 GPUs
RTX 5090: 603 GPUs
RTX 3090: 497 GPUs… |
@vast_ai |
Company |
Original |
2026-03-12 |
2,368 |
18 |
1 |
3 |
0 |
2 |
| Only 8 B200 GPUs left on Vast at $3.44/hr.
192GB HBM3e, ideal for fine-tuning.
Prices rise as supply tightens. |
@vast_ai |
Company |
Original |
2026-03-11 |
1,648 |
8 |
0 |
0 |
0 |
0 |
| GPU Market — Mar 11
H200: $2.45/hr (135 avail)
H100: $1.60/hr (36 avail)
B200: $3.44/hr (20 avail)
A100: $0.99/hr (44 … |
@vast_ai |
Company |
Original |
2026-03-11 |
1,323 |
8 |
0 |
0 |
0 |
2 |
| H200 on Vast: $2.45/hr
Same GPU on major clouds: $6.31+/hr
61% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-03-10 |
2,838 |
46 |
2 |
2 |
0 |
3 |
| Two of the largest corporate spend platforms in the US have independently named https://t.co/2WHvjspuyh one of their fa… |
@vast_ai |
Company |
Original |
2026-03-10 |
2,300 |
13 |
3 |
3 |
0 |
2 |
| 2,409+ GPUs available right now on Vast
Highest availability:
RTX 5090: 809 GPUs
RTX 4090: 757 GPUs
RTX 3090: 472 GPUs… |
@vast_ai |
Company |
Original |
2026-03-10 |
1,603 |
8 |
3 |
2 |
0 |
3 |
| Two independent corporate spend platforms now say the same thing:
Ramp: https://t.co/DfwOtzr83c Top 10 trending as a b… |
@vast_ai |
Company |
Original |
2026-03-10 |
955 |
2 |
0 |
0 |
0 |
0 |
| Only 17 B200 GPUs left on Vast at $3.41/hr.
192GB HBM3e, ideal for fine-tuning.
Prices rise as supply tightens. |
@vast_ai |
Company |
Original |
2026-03-09 |
1,703 |
18 |
0 |
0 |
0 |
0 |
| GPU Market — Mar 09
H200: $2.45/hr (120 avail)
H100: $1.60/hr (67 avail)
B200: $3.41/hr (37 avail)
A100: $0.99/hr (93 … |
@vast_ai |
Company |
Original |
2026-03-09 |
1,497 |
6 |
0 |
2 |
1 |
2 |
| H200 on Vast: $2.45/hr
Same GPU on major clouds: $6.31+/hr
61% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-03-08 |
1,874 |
8 |
0 |
1 |
0 |
0 |
| 2,666+ GPUs available right now on Vast
Highest availability:
RTX 4090: 885 GPUs
RTX 5090: 859 GPUs
RTX 3090: 480 GPUs… |
@vast_ai |
Company |
Original |
2026-03-08 |
3,629 |
68 |
2 |
7 |
0 |
5 |
| Only 44 B200 GPUs left on Vast at $3.13/hr.
192GB HBM3e, ideal for fine-tuning.
Prices rise as supply tightens. |
@vast_ai |
Company |
Original |
2026-03-07 |
3,605 |
26 |
4 |
0 |
0 |
3 |
| GPU Market — Mar 07
H200: $2.45/hr (199 avail)
H100: $1.60/hr (83 avail)
B200: $3.13/hr (37 avail)
A100: $0.95/hr (66 … |
@vast_ai |
Company |
Original |
2026-03-07 |
2,937 |
53 |
0 |
1 |
0 |
4 |
| H200 on Vast: $2.45/hr
Same GPU on major clouds: $6.31+/hr
61% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-03-06 |
3,225 |
39 |
1 |
5 |
0 |
3 |
| 2,231+ GPUs available right now on Vast
Highest availability:
RTX 5090: 652 GPUs
RTX 4090: 621 GPUs
RTX 3090: 532 GPUs… |
@vast_ai |
Company |
Original |
2026-03-06 |
2,012 |
10 |
2 |
0 |
0 |
2 |
| Only 16 B200 GPUs left on Vast at $3.06/hr.
192GB HBM3e, ideal for frontier model training.
Prices rise as supply tight… |
@vast_ai |
Company |
Original |
2026-03-05 |
1,828 |
3 |
0 |
0 |
0 |
2 |
| GPU Market — Mar 05
H200: $2.45/hr (136 avail)
H100: $1.60/hr (50 avail)
B200: $2.88/hr (21 avail)
A100: $0.93/hr (63 … |
@vast_ai |
Company |
Original |
2026-03-05 |
1,924 |
7 |
2 |
1 |
0 |
4 |
| H200 on Vast: $2.45/hr
Same GPU on major clouds: $6.31+/hr
61% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-03-04 |
1,919 |
4 |
1 |
1 |
0 |
1 |
| Two independent corporate spend platforms now say the same thing:
Ramp → Vast ai trending as a breakout growth vendor … |
@vast_ai |
Company |
Quote |
2026-03-04 |
2,856 |
6 |
3 |
1 |
0 |
5 |
| 2,112+ GPUs available right now on Vast
Highest availability:
RTX 5090: 816 GPUs
RTX 4090: 645 GPUs
RTX 3090: 366 GPUs… |
@vast_ai |
Company |
Original |
2026-03-04 |
1,234 |
5 |
0 |
3 |
0 |
0 |
| Only 25 B200 GPUs left on Vast at $2.94/hr.
192GB HBM3e, ideal for frontier model training.
Prices rise as supply tight… |
@vast_ai |
Company |
Original |
2026-03-03 |
2,148 |
9 |
1 |
2 |
0 |
1 |
| New vLLM-Omni template now available on https://t.co/srYVud1ciC.
Find it in the vast console: https://t.co/MYxRa3czaZ… |
@vast_ai |
Company |
Original |
2026-03-03 |
2,362 |
9 |
1 |
4 |
0 |
2 |
| GPU Market — Mar 03
H200: $2.45/hr (149 avail)
H100: $1.60/hr (76 avail)
B200: $2.94/hr (39 avail)
A100: $0.90/hr (50 … |
@vast_ai |
Company |
Original |
2026-03-03 |
1,969 |
7 |
1 |
2 |
1 |
1 |
| Evening GPU prices on Vast:
H200: $2.45/hr
H100: $1.60/hr
A100: $0.91/hr
RTX 5090: $0.41/hr
RTX 4090: $0.33/hr
Kick o… |
@vast_ai |
Company |
Original |
2026-03-03 |
1,823 |
4 |
0 |
1 |
0 |
0 |
| H200 on Vast: $2.45/hr
Same GPU on major clouds: $6.31+/hr
61% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-03-02 |
1,462 |
4 |
0 |
1 |
0 |
0 |
| Only 15 B200 GPUs left on Vast at $2.94/hr.
192GB HBM3e, ideal for frontier model training.
Prices rise as supply tight… |
@vast_ai |
Company |
Original |
2026-03-02 |
1,396 |
7 |
1 |
1 |
0 |
0 |
| 2,250+ GPUs available right now on Vast
Highest availability:
RTX 5090: 859 GPUs
RTX 4090: 743 GPUs
RTX 3090: 276 GPUs… |
@vast_ai |
Company |
Original |
2026-03-02 |
1,124 |
2 |
1 |
1 |
0 |
1 |
| RTX 4090 on Vast
From $0.33/hr — 768 available
24GB GDDR6X
Great for: Image generation, inference, fine-tuning
No con… |
@vast_ai |
Company |
Original |
2026-03-01 |
1,655 |
17 |
0 |
1 |
0 |
2 |
| RTX 5090 prices rose 2% since yesterday — now $0.42/hr on Vast.
786 available, 32GB GDDR7. No minimum commitment. |
@vast_ai |
Company |
Original |
2026-03-01 |
2,434 |
18 |
0 |
2 |
0 |
3 |
| GPU Market — Mar 01
H200: $2.45/hr (159 avail)
H100: $1.60/hr (57 avail)
B200: $2.94/hr (40 avail)
A100: $0.89/hr (79 … |
@vast_ai |
Company |
Original |
2026-03-01 |
3,729 |
32 |
2 |
3 |
1 |
5 |
| Evening GPU prices on Vast:
H200: $2.45/hr
H100: $1.60/hr
A100: $0.88/hr
RTX 5090: $0.41/hr
RTX 4090: $0.33/hr
Kick o… |
@vast_ai |
Company |
Original |
2026-03-01 |
160,261 |
158 |
2 |
10 |
2 |
215 |
| H200 on Vast: $2.45/hr
Same GPU on major clouds: $6.31+/hr
61% cheaper. Same hardware. No contracts. |
@vast_ai |
Company |
Original |
2026-02-28 |
5,120 |
58 |
3 |
3 |
0 |
12 |
| Only 43 B200 GPUs left on Vast at $2.94/hr.
192GB HBM3e, ideal for frontier model training.
Prices rise as supply tight… |
@vast_ai |
Company |
Original |
2026-02-28 |
4,908 |
32 |
3 |
4 |
0 |
10 |
| 2,803+ GPUs available right now on Vast
Highest availability:
RTX 5090: 908 GPUs
RTX 4090: 874 GPUs
RTX 3090: 516 GPUs… |
@vast_ai |
Company |
Original |
2026-02-28 |
2,736 |
33 |
2 |
2 |
0 |
6 |
| GPU Market — Feb 28
H200: $2.45/hr (160 avail)
H100: $1.60/hr (86 avail)
B200: $2.94/hr (75 avail)
A100: $0.88/hr (62 … |
@vast_ai |
Company |
Original |
2026-02-28 |
108,540 |
890 |
49 |
24 |
8 |
504 |
| H100 from $1.55/hr on https://t.co/2WHvjspuyh — 75 available right now.
80GB HBM3, no contracts, pay per second. |
@vast_ai |
Company |
Original |
2026-02-27 |
1,045 |
5 |
1 |
0 |
0 |
0 |
| https://t.co/t0xkhdw9sK |
@vast_ai |
Company |
Original |
2026-02-06 |
979 |
4 |
1 |
1 |
0 |
0 |
| New KIMI K2.5 template now available on https://t.co/srYVud1ciC.
Find it in the https://t.co/srYVud1ciC model library:… |
@vast_ai |
Company |
Original |
2026-01-29 |
1,135 |
6 |
0 |
0 |
0 |
0 |
| New SGLang template now available on https://t.co/srYVud1ciC.
Learn more here: https://t.co/toZz1x0KIO
__
@sgl_projec… |
@vast_ai |
Company |
Original |
2026-01-23 |
1,159 |
7 |
0 |
0 |
0 |
1 |
| LTX-2 Template now available. Find it in the all-new https://t.co/DfwOtzr83c Model Library.
@ltx_model is a DiT-based … |
@vast_ai |
Company |
Original |
2026-01-21 |
777 |
4 |
0 |
0 |
0 |
1 |
| Serverless GPU inference for up to 80% less than anyone else. That’s 2–5× more efficiency.
https://t.co/2WHvjspuyh Ser… |
@vast_ai |
Company |
Original |
2025-12-16 |
3,412 |
15 |
2 |
2 |
1 |
5 |
| HUGE NEWS! 📣 https://t.co/2WHvjsoWIJ is thrilled to announce the launch of our Serverless offering for GPU workloads, e… |
@vast_ai |
Company |
Original |
2025-12-10 |
559 |
5 |
1 |
0 |
1 |
0 |
| 2025 https://t.co/2WHvjspuyh Product Launch Event, live-streamed from SF at 7:00 PM PT tonight. You won't want to miss … |
@vast_ai |
Company |
Original |
2025-12-09 |
1,183 |
6 |
3 |
0 |
0 |
1 |
| 🚨 Something big is coming from https://t.co/5CuOa5eIHW next week.
We’re hosting a live product launch with an in-perso… |
@vast_ai |
Company |
Original |
2025-12-03 |
1,234 |
7 |
2 |
1 |
1 |
0 |
| some new signage at the LA office 😎 https://t.co/XeUfqx2Bxn |
@vast_ai |
Company |
Original |
2025-12-02 |
515 |
10 |
0 |
0 |
1 |
0 |
| Something big is coming...
Live from San Francisco, CA
🎥 𝟭𝟮 / 𝟬𝟵 / 𝟮𝟬𝟮𝟱 https://t.co/hElz9NDcso |
@vast_ai |
Company |
Original |
2025-11-28 |
1,167 |
5 |
2 |
0 |
0 |
2 |
| What if your Integrated Development Environment (IDE) could talk back to you, suggest solutions, and write half your co… |
@vast_ai |
Company |
Original |
2025-11-18 |
335 |
2 |
0 |
1 |
0 |
0 |
| When you think of a large language model (LLM), you probably picture the big household names like ChatGPT, Claude, Gemi… |
@vast_ai |
Company |
Original |
2025-11-14 |
375 |
1 |
0 |
0 |
0 |
0 |
| From containerized environments to our Secure Cloud offering and ongoing compliance efforts, here's how we help ensure … |
@vast_ai |
Company |
Original |
2025-11-04 |
441 |
7 |
0 |
1 |
0 |
0 |
| These past few months, we've expanded our offerings so you can get the most out of your time and budget – with new hard… |
@vast_ai |
Company |
Original |
2025-10-31 |
397 |
5 |
0 |
0 |
0 |
0 |
