Large Language Models
docAnalyzer keeps pace with releases from Alibaba, Anthropic, DeepSeek, Google, Minimax, Moonshot AI, OpenAI, Xiaomi, and xAI, adding new models within days so you can pick the right fit for every workflow. We surface 21 models across 9 creators, each benchmarked for quality, speed, and responsiveness.
Refreshed 2026-09-12
50.7 top quality · Claude Opus 5
345 tok/s top speed · Gemini 3.5 Flash Lite
0 ms top responsiveness · GPT-5 nano
| Model | |||
|---|---|---|---|
 Claude Opus 5 Anthropic | 50.7 | 61 tok/s | 36.03 s |
 GPT-5.6 Sol OpenAI | 47.1 | 69 tok/s | 124.83 s |
 Grok 4.6 xAI | 44.4 | 71 tok/s | 38.01 s |
 Kimi K3 Moonshot AI | 43.8 | 37 tok/s | 2.99 s |
| GPT-5.6 Terra OpenAI | 42.3 | 116 tok/s | 107.22 s |
 Gemini 3.8 Flash Google | 41.2 | 339 tok/s | 12.31 s |
 Qwen3.8 Max Alibaba | 40.3 | 41 tok/s | 1.80 s |
 DeepSeek V4.1 Flash DeepSeek | 39.5 | 261 tok/s | 920 ms |
| Claude Sonnet 5 Anthropic | 38.4 | 83 tok/s | 114.22 s |
| GPT-5.6 Luna OpenAI | 37.5 | 121 tok/s | 96.08 s |
| Qwen 3.8 27B Alibaba | 33.9 | 43 tok/s | 1.31 s |
| Gemini 3.1 Pro Preview Google | 30.4 | 125 tok/s | 24.48 s |
| Qwen3.7 Max Alibaba | 29.9 | 0 tok/s | 0 ms |
 MiniMax M3 Minimax | 29.6 | 106 tok/s | 830 ms |
 MiMo V2.5 Pro Xiaomi | 26.4 | 44 tok/s | 7.72 s |
| Qwen3.7 Plus Alibaba | 25.8 | 70 tok/s | 1.46 s |
| Grok 4.3 xAI | 25.4 | 0 tok/s | 0 ms |
| Gemini 3.5 Flash Lite Google | 22.7 | 345 tok/s | 7.71 s |
| MiMo V2.5 Xiaomi | 22.3 | 52 tok/s | 6.13 s |
| Claude Haiku 4.5 Anthropic | 17.6 | 106 tok/s | 12.43 s |
| GPT-5 nano OpenAI | 13.0 | 0 tok/s | 0 ms |
* Latency uses a logarithmic scale: shorter time means a longer bar.