AI Model Inference Providers
25 companies in this space
Important Developer Trends for December 2023
Reconstructed with current taxonomy
| Rank | Trend | Score | Change | Components | Recent scores |
|---|---|---|---|---|---|
| #1 |
LLM
LLMs are a core workload for AI inference providers, directly shaping model availability, API capabilities, pricing, performance, and developer platform selection across the market.
|
85.6 | — |
Strategic 85 Adoption 88 · Attention 100 · Momentum 41 |
77 · 82 · 89 · 89 · 86 · 84 · 88 · 86 |
| #2 |
AI Model Fine-tuning
Fine-tuning is a direct extension of inference offerings, letting developers customize model behavior, quality, cost, and deployment characteristics for production use cases.
|
79.4 | — |
Strategic 78 Adoption 72 · Attention 100 · Momentum 64 |
70 · 75 · 80 · 82 · 75 · 77 · 74 · 79 |
| #3 |
Real-time
Real-time response delivery and streaming outputs are important capabilities for interactive AI inference applications, affecting latency, user experience, and API integration choices.
|
69.6 | — |
Strategic 68 Adoption 56 · Attention 100 · Momentum 52 |
55 · 67 · 68 · 69 · 64 · 64 · 71 · 70 |
| #4 |
Serverless
Serverless and function-based deployment models directly affect how developers provision, scale, and pay for AI inference workloads.
|
64.6 | — |
Strategic 68 Adoption 32 · Attention 100 · Momentum 48 |
64 · 47 · 64 · 57 · 57 · 51 · 69 · 65 |
| #5 |
Vector Search
Vector search is a durable building block for retrieval-augmented AI applications, shaping model selection, embedding workflows, latency, and integration requirements.
|
62.6 | +1 |
Strategic 66 Adoption 32 · Attention 100 · Momentum 32 |
58 · 63 · 69 · 68 · 61 · 62 · 66 · 63 |
Filter Companies
Summary
| Company | Total Posts | Avg/Month |
|---|---|---|
| Google Cloud | 2,784 | 8.0 |
| Nebius | 239 | 6.6 |
| Prem AI | 277 | 5.8 |
| DigitalOcean | 843 | 4.7 |
| Together AI | 239 | 4.0 |
| Monster API | 184 | 3.8 |
| Anthropic | 276 | 3.8 |
| Venice | 125 | 3.5 |
| Fireworks AI | 179 | 3.0 |
| Baseten | 260 | 2.7 |
| Modular | 153 | 2.6 |
| Deepinfra | 141 | 2.4 |
| Fal | 120 | 1.7 |
| Featherless | 66 | 1.4 |
| Replicate | 125 | 1.3 |
| Lambda | 214 | 1.2 |
| Predibase | 94 | 1.1 |
| Modal | 76 | 1.1 |
| OpenPipe | 27 | 0.6 |
| BentoML | 42 | 0.4 |
| Clarifai | 422 | 0.0 |
| OpenAI | 21 | 0.0 |
| Hugging Face | 682 | 0.0 |
| Neptune.ai | 268 | 0.0 |
| Mistral AI | 0 | 0.0 |
Current Subscribers
| Company | Subscribers | Share |
|---|---|---|
| OpenAI | 2,010,000 | |
| Google Cloud | 1,430,000 | |
| Anthropic | 759,000 | |
| Hugging Face | 140,000 | |
| DigitalOcean | 60,200 | |
| Mistral AI | 15,800 | |
| Modular | 15,000 | |
| Lambda | 7,540 | |
| Nebius | 6,700 | |
| Together AI | 4,600 | |
| Replicate | 3,920 | |
| Venice | 3,750 | |
| Baseten | 2,610 | |
| Modal | 1,980 | |
| Fireworks AI | 747 | |
| Featherless | 31 |
Showing Hacker News posts with 50+ points since 2022
Companies in AI Model Inference Providers
Random Space| Company | Stage | Founded |
|---|---|---|
| Anthropic | H | 2021 |
| Baseten | E | 2019 |
| BentoML | acquired | 2019 |
| Clarifai | acquired | 2013 |
| Deepinfra | B | 2022 |
| DigitalOcean | Public | 2012 |
| Fal | D | 2021 |
| Featherless | A | 2023 |
| Fireworks AI | D | 2022 |
| Google Cloud | Public | 1998 |
| Hugging Face | D | 2016 |
| Lambda | E | 2012 |
| Mistral AI | C | 2023 |
| Modal | C | 2021 |
| Modular | B | 2022 |
| Monster API | pre-seed | 2023 |
| Nebius | Public | 2024 |
| Neptune.ai | acquired | 2017 |
| OpenAI | F | 2015 |
| OpenPipe | seed | 2023 |
| Predibase | acquired | 2020 |
| Prem AI | A | 2023 |
| Replicate | acquired | 2019 |
| Together AI | C | 2022 |
| Venice | A | 2024 |