Context window

Maximum context length in tokens as listed by the provider.

As of 24 Oct 2026, BenchLeader ranks Gemini 1206 first on the Context window board at 2.1M, ahead of Grok 4.20 (2M) and Grok 4.20 Multi-Agent (2M).

486 models · data as of 24 Oct 2026

1. 1Gemini 12062.1M
2. 2Grok 4.20 (thinking)2M
3. 3Grok 4.20 Multi-Agent2M
4. 4Grok 4 Fast (thinking)2M
5. 5Gemini 2.0 Pro2M
6. 6Gemini 1.5 Pro2M
7. 7Gemini 2.0 Flash Thinking (thinking)2M
8. 8GPT-6 Astra (max)1.1M
9. 9GPT-6.1 Sol (max)1.1M
10. 10GPT-5.6 Sol (max)1.1M
11. 11GPT-6 Sol (max)1.1M
12. 12GPT-5.5 (xhigh)1.1M
13. 13GPT-5.5 Pro (xhigh)1.1M
14. 14GPT-5.4 (xhigh)1.1M
15. 15GPT-5.6 Terra (max)1.1M

SearchAll providersAi2AI21 LabsAI9StarsAlibabaAmazonAnt GroupAnthropicApodexArcee AIBaiduByteDanceCelerisChina MobileCohereDeep CogitoDeepSeekGoogleIBMInceptionKTKuaishouKwaiKATLG AI ResearchLiquid AIMBZUAIMeituanMetaMicrosoftMiniMaxMistral AIMoonshot AIMotif TechnologiesMultiverse ComputingNanbeigeNaverNex AGINous ResearchNVIDIAOpenAIOpenBMBOpenChatPerplexityPoolsidePrime IntellectReflectionRekaSapiens AISarvamServiceNowSK TelecomSpaceXAIStepFunSwiss AITencentThinking MachinesTIITrillion LabsUnknownUpstageWriterXiaomiZhipu AIOpen weightsNew models486 of 486

#	Model	tokens ↓
1	Gemini 1206Google	2.1M
2	Grok 4.20thinkingSpaceXAI	2M
3	Grok 4.20 Multi-AgentSpaceXAI	2M
4	Grok 4 FastthinkingSpaceXAI	2M
5	Gemini 2.0 ProGoogle	2M
6	Gemini 1.5 ProGoogle	2M
7	Gemini 2.0 Flash ThinkingthinkingGoogle	2M
8	GPT-6 AstramaxOpenAI	1.1M
9	GPT-6.1 SolmaxOpenAI	1.1M
10	GPT-5.6 SolmaxOpenAI	1.1M
11	GPT-6 SolmaxOpenAI	1.1M
12	GPT-5.5xhighOpenAI	1.1M
13	GPT-5.5 ProxhighOpenAI	1.1M
14	GPT-5.4xhighOpenAI	1.1M
15	GPT-5.6 TerramaxOpenAI	1.1M
16	GPT-5.4 ProOpenAI	1.1M
17	GPT-6 LunamaxOpenAI	1.1M
18	GPT-5.6 LunamaxOpenAI	1.1M
19	GPT-5.6 Sol PromaxOpenAI	1.1M
20	GPT-5.6OpenAI	1.1M
21	GPT-6 Astra PromaxOpenAI	1.1M
22	LongCat 2.0Meituanopen ↗	1.0M
23	Gemini 4 ArgonhighGoogle	1.0M
24	Muse WondermaxMeta	1.0M
25	Muse Spark 1.3maxMeta	1.0M
26	NewKimi K3.1 ProMoonshot AI	1.0M
27	Kimi K3maxMoonshot AIopen ↗	1.0M
28	Gemini 3.8 FlashhighGoogle	1.0M
29	MiMo-V2.6-ProXiaomiopen ↗	1.0M
30	NewDeepSeek V4.1 Flash 1017maxDeepSeekopen ↗	1.0M
31	Gemini 3.7 FlashhighGoogle	1.0M
32	Muse Spark 1.2xhighMeta	1.0M
33	Gemini 3.5 FlashhighGoogle	1.0M
34	Muse Spark 1.1Meta	1.0M
35	Gemini 3.6 FlashhighGoogle	1.0M
36	Muse Spark OpenxhighMetaopen ↗	1.0M
37	DeepSeek V4.1 FlashmaxDeepSeekopen ↗	1.0M
38	DeepSeek V4 PromaxDeepSeekopen ↗	1.0M
39	Gemini 3.1 ProGoogle	1.0M
40	MiMo-V2.6-FlashXiaomiopen ↗	1.0M
41	DeepSeek V4 FlashmaxDeepSeekopen ↗	1.0M
42	Gemini 3 FlashthinkingGoogle	1.0M
43	NewMuse Spark Open 816B A42BxhighMetaopen	1.0M
44	Deepseek v4 Flash VisionmaxDeepSeek	1.0M
45	MiMo-V2.5-ProthinkingXiaomiopen ↗	1.0M
46	Mistral Large 4Mistral AIopen	1.0M
47	Gemini 3.5 Flash LiteGoogle	1.0M
48	NewBeam 1.0 501B A23BReflectionopen ↗	1.0M
49	Gemini 2.5 ProGoogle	1.0M
50	Gemini 3.1 Flash LiteGoogle	1.0M
51	Nemotron 3.5 Super VL 120B A12BNVIDIAopen	1.0M
52	Solar Open2 250BUpstageopen ↗	1.0M
53	Gemini 2.5 FlashGoogle	1.0M
54	NewBeam 1.0Reflectionopen	1.0M
55	Gemini 2.5 Flash-LiteGoogle	1.0M
56	Gemini 1.5 Flash 8BGoogle	1.0M
57	Gemini Pro LatestGoogle	1.0M
58	Nemotron 3.5 Super VLNVIDIAopen ↗	1.0M
59	UEmber-1Unknown	1.0M
60	Gemini 3.1 Pro CustomtoolsGoogle	1.0M

Show 100 more

Cite as: BenchLeader, “Context window”, data as of 24 Oct 2026, https://www.benchleader.com/leaderboards/context.