Live index · 49 models · Updated Oct 2026
RSS/API/DOCS

ModelRegistry

v2026.9
Add

Reference · updated 2026-10-07 (dataset 2026.10.07-r2)

Context window comparison

Maximum input tokens per model, largest first. A dash means undisclosed or non-token (audio/video billing) - never zero. Long context alone says nothing about retrieval quality; per-record citations are on each model page.

01Llama 4 Scout (16E)1.31MSource
02MiMo-V2.51.10MSource
03MiMo-V2.5 Pro1.10MSource
04GPT-6 Astra1.05MSource
05ChatGPT 5.61.05MSource
06GPT-6.1 Sol1.05MSource
07GPT-6 Luna1.05MSource
08Gemini 4 Argon1.05MSource
09Gemini 3.8 Flash1.05MSource
10DeepSeek V4-Pro (0813)1.05MSource
11Llama 4 Maverick (128E)1.05MSource
12Hy4 Preview1.05MSource
13GLM 5.31.05MSource
14GLM 5.3 Flash1.05MSource
15GLM 5.21.05MSource
16MiniMax M31.05MSource
17DeepSeek V4 Flash 07311.05MSource
18DeepSeek V4 Flash 04231.05MSource
19Gemini 3.7 Flash1.05MSource
20Claude Fable 5.11MSource
21Claude Opus 51MSource
22Claude Opus 5.51MSource
23DeepSeek V4.1 Flash1MSource
24Qwen3.8 2.4T A95B1MSource
25Qwen3.8 Flash1MSource
26Qwen3.8 Omni Flash1MSource
27Mistral Large 41MSource
28MiMo-V2.6 Pro1MSource
29MiMo-V2.6 Flash1MSource
30Claude Sonnet 51MSource
31Claude Sonnet 5.51MSource
32Grok 4.6500KSource
33Grok 4.7500KSource
34DeepSeek V4 Flash Vision Exp262KSource
35Muse Spark 1.3262KSource
36Mistral Medium 3.5262KSource
37Hy3262KSource
38Nemotron 3 Ultra 550B262KSource
39Kimi K3262KSource
40Gemini 3.8 Live Extended Thinking131KSource
41Sarvam 105B128KSource
42Nano Banana Pro66KSource
43Grok Voice Transcribe 2.0-Source
44Muse Voice Transcribe-Source
45ChatGPT Images 2.5-Source
46Veo 3.1-Source
47Kling 3.0-Source
48Runway Gen-4.5-Source
49Jev-Source