modelbenchmark.io

Best AI for long context

Models with a listed context window of 128K tokens or more. Ranked by composite when scored, else by context size. Full leaderboard.

RankModelLabCompositeContextOut $/M
1GPT 6 AstraOpenAI100th1.1M$50.00
2Claude Fable 5Anthropic99th1M$50.00
3Claude Fable 5.1Anthropic99th1M$50.00
4GPT-5.5 ProOpenAI98th1.1M$180.00
5Claude Opus 5Anthropic96th1M$25.00
6GPT-5.4 ProOpenAI95th1.1M$180.00
7Gemini 3.8 FlashGoogle95th1M$3.75
8GPT-5.6 SolOpenAI94th1.1M$10.00
9DeepSeek V4.1 FlashDeepSeek94th1M$1.2778
10Gemini 3.7 FlashGoogle94th1M$3.75
11GPT 5.5OpenAI93rd1.1M$30.00
12Gemini-3-ProGoogle91st1M$9.60
13MiniMax-M2.5MiniMax91st205K$1.20
14GPT 5.6 TerraOpenAI90th1.1M$12.00
15Qwen3.8 Max ThinkingAlibaba89th991K$6.00
16Claude Opus 4.8Anthropic89th1M$25.00
17GPT 5.3 CodexOpenAI89th400K$14.00
18GPT 5.4OpenAI88th1.1M$15.00
19Grok 4.6xAI87th500K$6.00
20Gemini 3.1 Pro (Preview)Google86th1M$12.00
21Grok 4.5xAI85th500K$6.00
22Gemini 3 Flash (Preview)Google84th1M$3.00
23Gemini 3.6 FlashGoogle84th1M$3.75
24Gemini 3.5 FlashGoogle84th1M$9.00
25Qwen3.7 PlusAlibaba84th1M$3.072
26Claude 4.7 OpusAnthropic83rd1M$25.00
27GLM-5.3Zhipu82nd1M$4.40
28Claude Sonnet 5 ThinkingAnthropic80th1M$10.00
29GPT-5.6 LunaOpenAI80th1.1M$0.37
30DeepSeek V4 Flash Vision ExpDeepSeek80th1M$1.32
31OpenAI/GPT-5.2OpenAI79th400K$14.00
32Qwen3.7 MaxAlibaba79th1M$9.00
33GPT 5 ProOpenAI78th400K$120.00
34DeepSeek-R1DeepSeek78th128K$1.70
35OpenAI/GPT-5OpenAI78th400K$10.00
36GLM-5.1Zhipu77th200K$4.40
37GPT 5.1 CodexOpenAI77th400K$10.00
38OpenAI o3OpenAI76th200K$8.00
39GPT-5.1 (2025-11-13)OpenAI75th400K$10.00
40Claude 4.6 Opus ThinkingAnthropic74th1M$25.00
41GPT 5.2 CodexOpenAI73rd400K$14.00
42Qwen3.6 35B-A3BAlibaba73rd262K$2.052
43Kimi K2Kimi72nd128Kfree
44Kimi K2.6Kimi72nd262K$4.00
45Moonshotai/Kimi-K2.5Alibaba71st256K$1.90
46Qwen3.5 397B-A17BAlibaba70th262K$2.65
47DeepSeek V3.2DeepSeek68th128K$0.326
48Minimax/Minimax-M2MiniMax68th200K$1.53
49GLM-5Alibaba66th205K$3.2647
50GLM-5.2Alibaba66th1M$4.40
51Qwen3.5 Plus ThinkingAlibaba65th984K$2.40
52Kimi K2.7 CodeKimi63rd262K$4.389
53Qwen3.5 35B A3BAlibaba63rd260K$1.80
54GLM-5.3-FlashZhipu61st1M$0.025
55Inkling ThinkingDatabricks61st1M$4.05
56OpenAI o1OpenAI59th200K$60.00
57Qwen3 Max PreviewAlibaba58th256K$6.001
58Gemini 2.5 ProGoogle57th1M$10.00
59Gemini 3.1 Flash LiteGoogle56th1M$1.50
60Claude 4.5 OpusAnthropic55th200K$25.00
61OpenAI o4-miniOpenAI55th200K$4.40
62GPT 5 MiniOpenAI54th400K$2.00
63Claude Sonnet 4.5Anthropic54th200K$15.00
64Z-Ai/GLM 4.7Zhipu51st200K$0.80
65GPT 5.4 MiniOpenAI50th400K$4.50
66Qwen3.5 9BAlibaba48th256K$0.15
67GPT OSS 120BOpenAI47th131K$0.798
68GPT 5.4 NanoOpenAI46th400K$1.25
69Qwen3.6 FlashAlibaba46th992K$1.16
70Qwen3.5 122B A10B ThinkingAlibaba45th131K$3.496
71Claude 4.1 OpusAnthropic43rd200K$75.00
72GLM 4.7 Flash ThinkingZhipu43rd200K$0.40
73OpenAI o3-miniOpenAI42nd200K$4.40
74GPT 4.1OpenAI41st1M$8.00
75Gemini 2.5 FlashGoogle40th1M$2.50
76GPT 5 NanoOpenAI40th400K$0.40
77Gemini 3.5 Flash LiteGoogle39th1M$2.50
78Z-AI/GLM 4.6Zhipu38th200K$1.40
79DeepSeek-V3DeepSeek35th128K$0.77
80DeepSeek V4 ProDeepSeek35th1M$0.87
81GLM 4.5Zhipu34th131K$1.30
82GPT 4.1 MiniOpenAI31st1M$1.60
83GPT 4.1 NanoOpenAI22nd1M$0.40
84GPT-4o (2024-08-06)OpenAI21st128K$10.00
85Mistral Large 2411Mistral15th128K$6.001
86GPT-4o miniOpenAI15th128K$0.60
87Grok Build 0.1xAI12th256K$2.00
88Grok 4.3xAI0th1M$2.50
89X-Ai/Grok 4.1 Fast Non ReasoningxAI2M$0.50
90Gemini 2.5 Flash LiteGoogle1M$0.40
91Kimi K3Alibaba1M$15.00
92MiMo-V2.51M$2.00
93MiMo-V2.5-Pro1Mfree
94Claude 4 Sonnet ThinkingAnthropic1M$15.00
95Claude Sonnet 4.6Anthropic1M$15.00
96DeepSeek V4 FlashDeepSeek1M$0.196
97Qwen 3.6 PlusAlibaba992K$1.95
98Qwen3.8 FlashAlibaba992K$0.42
99MiniMax M3 ThinkingMiniMax512K$1.20
100GPT 5.1 Codex MaxOpenAI400K$20.00
101GPT 5.1 Codex MiniOpenAI400K$2.00
102Gemma 4 26B A4BGoogle262K$0.38
103Gemma 4 31B IT (free)Google262Kfree
104Qwen3 235B A22B Instruct 2507Alibaba262K$3.078
105Qwen3 235B A22B Thinking 2507Alibaba262K$0.60
106Qwen3 Coder NextAlibaba262K$1.50
107Qwen3.8 27BAlibaba262K$0.70
108Step 3.7 Flash262K$1.14
109Tencent Hy3262K$0.26
110Qwen3 Coder 480B A35B InstructAlibaba262K$1.00
111Qwen3.5 27BAlibaba260K$2.16
112Qwen3.6 27B ThinkingAlibaba260K$2.24
113Codestral 2508Mistral256K$0.90
114Kimi K2 ThinkingKimi256K$2.50
115MiniMax M2.7MiniMax205K$1.26
116MiniMax-M2.7-highspeedMiniMax205K$2.40
117Minimax/Minimax-M2.1MiniMax205K$1.32
118GLM 5 TurboZhipu203K$4.00
119GLM 5V Turbo ThinkingZhipu203K$4.00
120Claude 3.7 SonnetAnthropic200K$15.00
121Claude 4 Opus ThinkingAnthropic200K$75.00
122Claude Haiku 4.5 ThinkingAnthropic200K$5.00
123DeepSeek V3.2 ExpDeepSeek164K$0.42
124Llama 3.1 8b InstructMeta131K$0.085
125Llama 3.2 3b InstructMeta131K$0.0493
126Llama 4 Maverick 17B 128E Instruct FP8Meta131K$1.484
127Qwen3 Next 80B A3B InstructAlibaba131K$0.65
128Qwen3 Next 80B A3B ThinkingAlibaba131K$0.65
129Qwen3 VL 235B A22B InstructAlibaba131K$1.20
130Qwen3 VL 235B A22B ThinkingAlibaba131K$6.00
131GLM 4.5 AirZhipu131K$0.80
132Cerebras-Llama-4-Scout-17B-16E-InstructMeta128Kfree
133DeepSeek-V3.1DeepSeek128K$0.70
134GPT-4 TurboOpenAI128K$30.00
135gpt-oss-20bOpenAI128K$0.15
136Mistral Small 3.2 (Mistral AI)Mistral128K$0.30
137Qwen 3 235B A22BAlibaba128K$0.50
138Qwen3 Coder FlashAlibaba128K$1.50
139Qwen3 Coder PlusAlibaba128K$5.00
140Qwen3-Coder 30B-A3B InstructAlibaba128K$1.083

Common questions

What is the best ai for long context right now?
Models with a listed context window of 128K tokens or more. Ranked by composite when scored, else by context size.
How is this ranking computed?
The set is a spec filter on the catalog. Scored models rank by the overall composite. Unscored models follow by the spec itself.