Large Language Models

docAnalyzer keeps pace with releases from Alibaba, Anthropic, Google, Moonshot AI, OpenAI, Zhipu AI, and xAI, adding new models within days so you can pick the right fit for every workflow instead of being locked to a single AI source. We currently surface 13 models across 7 creators, each benchmarked for quality and speed so you can choose confidently. Explore the latest line-up below.

Model	Quality	Speed	Latency*
Gemini 3.1 Pro Preview Google	57 57	122 122 token/s	44 44 seconds
GPT‑5.4 OpenAI	57 57	70 70 token/s	148.27 148.27 seconds
Claude Sonnet 4.6 Anthropic	44 44	45 45 token/s	1.63 1.63 seconds
GPT-5 mini OpenAI	41 41	78 78 token/s	98.63 98.63 seconds
Z.ai GLM 5 Zhipu AI	41 41	68 68 token/s	1.39 1.39 seconds
Qwen3.5-397B-A17B Alibaba	40 40	85 85 token/s	2.69 2.69 seconds
Grok 4.1 Fast xAI	39 39	151 151 token/s	8.49 8.49 seconds
Claude Haiku 4.5 Anthropic	37 37	105 105 token/s	33.4 33.4 seconds
Qwen3.5-27B Alibaba	37 37	93 93 token/s	5.64 5.64 seconds
Kimi K2.5 Moonshot AI	37 37	46 46 token/s	2.9 2.9 seconds
Gemini 3.1 Flash Lite Preview Google	34 34	271 271 token/s	6.89 6.89 seconds
GPT OSS 120b OpenAI	33 33	279 279 token/s	0.82 0.82 seconds
GPT-5 nano OpenAI	27 27	138 138 token/s	110.84 110.84 seconds

* Latency progress bars use a logarithmic scale, where lower latency values indicate better performance.