Provider model counts in 2026
Model counts across 40 providers in the index. Counts include every variant and generation tracked in the dataset.
What the data shows
OpenAI has the largest lineup with 40 models. The median provider has 14 models.
A large lineup is not automatically better. It can mean more price points and more ways to match a model to a task, or it can mean more variants that are hard to tell apart.
How to read the ranking
- Compare the best model, not just the count.
- Check the average price for the tier you need.
- Look at the open-weight share if self-hosting matters.
- Keep a second provider for fallback.
Provider model counts
| # | Provider | Models | Avg input | Avg output | Best model |
|---|---|---|---|---|---|
| 1 | OpenAI | 40 | $16.38 | $73.38 | gpt-5-2-pro |
| 2 | Mistral AI | 35 | $2.71 | $12.15 | mistral-large-3 |
| 3 | Anthropic | 32 | $15.19 | $68.66 | claude-opus-4-5-pro |
| 4 | Google DeepMind | 31 | $12.46 | $57.05 | gemini-3-pro-max |
| 5 | Alibaba Cloud | 28 | $6.25 | $27.15 | qwen3-max-pro |
| 6 | Meta AI | 28 | $3.00 | $13.44 | llama-4-maverick-instruct |
| 7 | Cohere | 21 | $6.25 | $27.88 | command-a-pro |
| 8 | DeepSeek | 21 | $3.53 | $16.26 | deepseek-v3-2 |
| 9 | 01.AI | 14 | $2.39 | $10.97 | yi-lightning |
| 10 | Allen Institute for AI | 14 | $1.67 | $7.05 | olmo-2-instruct |
| 11 | Amazon | 14 | $9.72 | $43.75 | nova-pro-max |
| 12 | Baidu | 14 | $11.20 | $48.78 | ernie-4-5-max |
| 13 | ByteDance | 14 | $7.26 | $31.79 | doubao-1-5-pro-reasoning |
| 14 | EleutherAI | 14 | $0.83 | $3.56 | gpt-neox-instruct |
| 15 | Fireworks AI | 14 | $6.84 | $31.35 | firefunction-max |
| 16 | Groq | 14 | $2.30 | $10.23 | llama-3-3-groq |
| 17 | Hugging Face | 14 | $0.96 | $4.63 | smollm3-instruct |
| 18 | Microsoft | 14 | $2.04 | $9.46 | phi-4 |
| 19 | MiniMax | 14 | $3.09 | $14.12 | minimax-m2 |
| 20 | Moonshot AI | 14 | $3.59 | $15.38 | kimi-k2 |
| 21 | NVIDIA | 14 | $1.98 | $8.48 | nemotron-4 |
| 22 | OpenRouter | 14 | $7.07 | $32.01 | fusion-max |
| 23 | Perplexity | 14 | $10.58 | $48.10 | sonar-pro-max |
| 24 | Reka AI | 14 | $10.01 | $46.74 | reka-core-pro |
| 25 | Tencent | 14 | $6.38 | $27.35 | hunyuan-turbo-max |
| 26 | Together AI | 14 | $2.28 | $10.11 | llama-3-1-turbo |
| 27 | Zhipu AI | 14 | $3.21 | $14.64 | glm-4-6-instruct |
| 28 | xAI | 14 | $15.40 | $71.89 | grok-4-reasoning |
| 29 | Anyscale | 7 | $2.45 | $10.46 | llama-3-anyscale-instruct |
| 30 | Baseten | 7 | $2.36 | $10.92 | llama-3-baseten-instruct |
| 31 | Cerebras | 7 | $2.57 | $11.47 | llama-3-1-cerebras-instruct |
| 32 | Databricks | 7 | $2.00 | $9.73 | dbrx |
| 33 | DeepInfra | 7 | $2.38 | $10.83 | llama-3-deepinfra-instruct |
| 34 | Hyperbolic | 7 | $2.37 | $10.45 | llama-3-hyperbolic-instruct |
| 35 | Lambda | 7 | $2.44 | $10.93 | llama-3-lambda-instruct |
| 36 | Modal | 7 | $2.39 | $10.89 | llama-3-modal-instruct |
| 37 | Novita AI | 7 | $2.36 | $11.51 | llama-3-novita-instruct |
| 38 | Replicate | 7 | $2.40 | $11.15 | llama-3-replicate |
| 39 | Snowflake | 7 | $1.69 | $7.51 | arctic-instruct |
| 40 | Stability AI | 7 | $1.13 | $5.32 | stablelm-2-instruct |
Methodology
This story uses the AI & LLM Index dataset: 600 models, 40 providers and 12 benchmarks. Prices are listed per million tokens. Scores are modelled estimates on a 0–100 scale and are not official benchmark results. Figures are rounded for readability.
Related
FAQ
What does the Provider model counts in 2026 data show?
Model counts across 40 providers in the index. Counts include every variant and generation tracked in the dataset.
Are the numbers official?
No. Prices and scores are compiled for this demo index. Benchmark values are modelled estimates, not official results. Verify figures with the provider or benchmark owner before making decisions.
How should I use this ranking?
Use it to form a shortlist. Check the specific model pages, compare price and context, and test the shortlist on your own examples before committing.