llmboard.ai
Benchmarks
CompareRankings
llmboard.ai
Benchmarks
CompareRankings
HomeLeaderboardsLLM Leaderboard

Published capability score

LLM Leaderboard

Best LLMs Ranked by Capability

Method: llmboard-score-v1.0. Data as of 2026-08-07

Unique models170
RulePublished score snapshot
MetricOverall
Highest value96.5

On this page

  • Ranking
  • Highlights
  • FAQ

Published ranking

Core capability and specification columns are visible by default. Up to ten featured benchmarks are shown; every other published benchmark remains available in Columns.

170 rows
Columns

Show columns

1OPGPT-5.6-SolOpenAIGPT-5.6 Sol96.599.896.5100.086.2100.0100.0%1.1MClosed weights$5$3094.6%N/AN/AN/AN/AN/AN/AN/A90.4%N/A
2ACQwen3.7 MaxAlibaba Cloud / Qwen TeamQwen3.7 Max90.391.8N/A93.183.893.080.0%1MClosed weights$2.5$7.592.4%89.6%N/AN/AN/AN/AN/A94.3%N/AN/A
3BYSeed 2.1 ProByteDanceSeed 2.1 Pro87.4N/A87.7N/A84.490.960.0%N/AClosed weightsN/AN/AN/AN/AN/AN/AN/AN/AN/AN/A86.2%N/A
4OPGPT-5.6-TerraOpenAIGPT-5.6 Terra85.496.193.093.858.382.5100.0%1.1MClosed weights$2$1292.9%N/AN/AN/AN/AN/AN/AN/A87.5%N/A
5GOGemini 3.1 ProGoogleGemini 3.1 Pro84.399.186.0N/A78.565.280.0%1MClosed weights$2$1294.3%N/AN/AN/AN/AN/AN/AN/A85.9%N/A
6ANClaude Opus 4.8AnthropicClaude Opus 4.882.998.178.9N/A84.960.080.0%1MClosed weights$5$2593.6%N/AN/AN/AN/AN/AN/AN/A84.3%N/A
7MAKimi K2.6Moonshot AIKimi K2.682.892.289.561.976.893.8100.0%262.1KOpen weights$0.95$490.5%N/AN/AN/AN/AN/AN/AN/A86.3%N/A
8GOGemini 3 ProGoogleGemini 3 Pro82.293.5N/AN/A72.776.060.0%1MClosed weightsN/AN/A91.9%N/A100.0%N/AN/AN/AN/AN/AN/AN/A
9BYSeed 2.1 TurboByteDanceSeed 2.1 Turbo80.5N/A84.2N/A73.784.860.0%N/AClosed weightsN/AN/AN/AN/AN/AN/AN/AN/AN/AN/A84.9%N/A
10ZAGLM 5.2Zhipu AIGLM-5.280.392.7N/A84.745.7100.080.0%1MOpen weightsN/AN/A91.2%N/AN/AN/AN/AN/AN/AN/AN/AN/A
11TEHy3TencentHy380.191.477.286.163.0N/A80.0%256KOpen weightsN/AN/A90.4%N/AN/AN/AN/AN/AN/AN/A84.2%N/A
12DEDeepSeek-V4-ProDeepSeekDeepSeek-V4-Pro-Max80.190.171.983.965.289.2100.0%1MClosed weightsN/AN/A90.1%87.5%N/AN/A93.5%N/AN/AN/A83.4%N/A
13ANClaude Opus 4.7AnthropicClaude Opus 4.778.498.765.8N/A55.891.780.0%1MClosed weights$5$2594.2%N/AN/AN/AN/AN/AN/AN/A79.3%N/A
14GOGemini 2.5 ProGoogleGemini 2.5 Pro Preview 06-0578.081.0N/AN/A82.367.160.0%1MClosed weightsN/AN/A86.4%N/A88.0%N/A69.0%N/AN/AN/AN/AN/A
15OPGPT-5.6-LunaOpenAIGPT-5.6 Luna77.294.070.287.556.472.5100.0%1.1MClosed weights$0.20$1.292.3%N/AN/AN/AN/AN/AN/AN/A83.3%N/A
16OPGPT-5.5OpenAIGPT-5.576.898.180.762.571.262.5100.0%1.1MClosed weights$5$3093.6%N/AN/AN/AN/AN/AN/AN/A84.4%N/A
17OPGPT-5.4OpenAIGPT-5.475.695.768.481.353.973.3100.0%1.1MClosed weights$2.5$1592.8%N/AN/AN/AN/AN/AN/AN/A82.7%N/A
18DEDeepSeek-V3.2DeepSeekDeepSeek-V3.2-Speciale75.5N/AN/A93.851.683.260.0%131.1KClosed weightsN/AN/AN/AN/A96.0%N/AN/AN/AN/AN/AN/AN/A
19ACQwen3.5 397B A17BAlibaba Cloud / Qwen TeamQwen3.5-397B-A17B75.294.175.054.568.979.9100.0%262.1KOpen weights$0.43$2.5888.4%87.8%N/AN/AN/AN/AN/A92.6%69.0%N/A
20AMNova 2 ProAmazonNova 2 Pro75.167.0N/AN/A80.681.460.0%N/AClosed weightsN/AN/A81.4%81.6%92.3%N/A74.6%N/AN/AN/AN/AN/A
21GOGemma 4 31BGoogleGemma 4 31B74.375.0N/AN/A71.776.560.0%262.1KOpen weightsN/AN/A84.3%85.2%N/AN/AN/AN/AN/AN/AN/AN/A
22ANClaude Sonnet 3.5AnthropicClaude 3.5 Sonnet73.441.4N/A84.996.980.480.0%200KClosed weightsN/AN/A67.2%77.6%N/A90.4%N/A78.3%93.7%N/AN/AN/A
23ACQwen3.7 PlusAlibaba Cloud / Qwen TeamQwen3.7-Plus73.184.1N/A66.954.088.280.0%1MClosed weights$0.50$390.3%88.5%N/AN/AN/AN/AN/A94.6%N/AN/A
24GOGemini 3 FlashGoogleGemini 3 Flash72.691.4N/AN/A51.968.860.0%1MClosed weights$0.50$390.4%N/A99.7%N/AN/AN/AN/AN/AN/AN/A
25ACQwen3.6 PlusAlibaba Cloud / Qwen TeamQwen3.6 Plus72.479.0N/A67.462.880.980.0%1MClosed weights$0.50$390.4%88.5%N/AN/AN/AN/AN/A94.3%N/AN/A
26ACQwen3.6 27BAlibaba Cloud / Qwen TeamQwen3.6-27B72.284.5N/A45.580.875.980.0%262.1KOpen weightsN/AN/A87.8%86.2%N/AN/AN/AN/AN/AN/AN/AN/A
27NVNemotron 3 UltraNVIDIANemotron 3 Ultra (550B A55B)72.182.515.8100.076.187.4100.0%1MOpen weightsN/AN/A87.0%86.8%N/AN/AN/AN/AN/AN/A44.4%N/A
28DEDeepSeek-V4-FlashDeepSeekDeepSeek-V4-Flash-Max71.786.452.676.168.371.5100.0%1MClosed weightsN/AN/A88.1%86.2%N/AN/A91.6%N/AN/AN/A73.2%N/A
29MAKimi K2.5Moonshot AIKimi K2.571.684.156.158.970.689.7100.0%262.1KOpen weights$0.60$387.6%87.1%96.1%N/AN/AN/AN/AN/A74.9%N/A
30OPGPT-5OpenAIGPT-571.679.333.361.597.788.4100.0%400KClosed weights$1.25$1085.7%N/A94.6%92.5%N/A84.7%93.4%N/A54.9%N/A
31MELlama 3.1 405BMetaLlama 3.1 405B Instruct69.961.2N/A76.587.752.080.0%128KClosed weightsN/AN/A50.7%73.3%N/A87.3%N/A73.8%89.0%88.6%N/AN/A
32DEDeepSeek-V2.5DeepSeekDeepSeek-V2.569.5N/AN/A69.475.462.060.0%8.2KClosed weightsN/AN/AN/AN/AN/A80.4%N/A74.7%89.0%N/AN/AN/A
33ANClaude Opus 4.6AnthropicClaude Opus 4.669.293.175.4N/A24.181.380.0%1MClosed weights$5$2591.3%N/A99.8%N/AN/AN/AN/AN/A84.0%N/A
34OPGPT-5.2OpenAIGPT-5.269.194.647.485.924.492.9100.0%400KClosed weights$1.75$1492.4%N/A100.0%N/AN/AN/AN/AN/A65.8%N/A
35XIMiMo V2.5 ProXiaomiMiMo-V2.5-Pro68.970.088.291.819.676.4100.0%1MOpen weights$0.435$0.8766.7%68.5%N/A89.4%N/A86.2%N/AN/AN/AN/A
36GOGemma 4 26B A4BGoogleGemma 4 26B-A4B67.268.8N/AN/A67.564.260.0%262.1KOpen weightsN/AN/A82.3%82.6%N/AN/AN/AN/AN/AN/AN/AN/A
37XAGrok 4 FastxAIGrok 4 Fast66.279.317.564.188.982.1100.0%2MClosed weightsN/AN/A85.7%N/A92.0%N/A80.0%N/AN/AN/A44.9%N/A
38ZAGLM 4.7Zhipu AIGLM-4.766.279.349.138.982.781.4100.0%204.8KOpen weightsN/AN/A85.7%84.3%95.7%N/AN/AN/AN/AN/A52.0%N/A
39ACQwen3 235B A22B ThinkingAlibaba Cloud / Qwen TeamQwen3-235B-A22B-Thinking-250765.766.4N/A58.361.679.980.0%262.1KClosed weightsN/AN/A81.1%84.4%92.3%N/AN/AN/AN/A87.8%N/AN/A
40ZAGLM 5.1Zhipu AIGLM-5.164.780.665.859.534.184.4100.0%200KOpen weightsN/AN/A86.2%N/AN/AN/AN/AN/AN/AN/A79.3%N/A
41ANClaude Sonnet 3.7AnthropicClaude 3.7 Sonnet63.676.7N/A56.5N/A51.360.0%200KClosed weightsN/AN/A84.8%N/A54.8%N/AN/AN/AN/A93.2%N/A80.0%
42GOGemini 2.5 FlashGoogleGemini 2.5 Flash63.570.9N/AN/A56.360.860.0%1MClosed weights$0.30$2.582.8%N/A72.0%N/AN/AN/AN/AN/AN/A88.0%
43AMNova ProAmazonNova Pro63.555.0N/A74.252.777.980.0%300KClosed weights$0.80$3.246.9%N/AN/A85.9%N/A76.6%89.0%92.1%N/AN/A
44MIMAI Thinking 1MicrosoftMAI-Thinking-163.474.4N/A30.077.870.380.0%N/AClosed weightsN/AN/A84.2%85.0%97.0%N/AN/AN/AN/AN/AN/AN/A
45OPo1OpenAIo163.059.7N/A64.862.367.080.0%200KClosed weights$15$6078.0%N/AN/A91.8%N/A96.4%88.1%N/AN/A74.3%
46NVLlama 3.3 Nemotron Super 49BNVIDIALlama-3.3 Nemotron Super 49B v162.639.7N/A64.596.952.580.0%131.1KOpen weightsN/AN/A66.7%N/A58.4%N/AN/AN/AN/AN/AN/AN/A
47NVLlama 3.1 Nemotron Ultra 253BNVIDIALlama 3.1 Nemotron Ultra 253B v162.657.3N/A75.866.748.180.0%N/AClosed weightsN/AN/A76.0%N/A72.5%N/A66.3%N/AN/A89.5%N/AN/A
48ZAGLM 4.5Zhipu AIGLM-4.562.061.63.593.572.584.5100.0%131.1KOpen weightsN/AN/A79.1%84.6%N/AN/A72.9%N/AN/AN/A26.4%91.0%
49OPo3 miniOpenAIo3-mini61.658.4N/A56.369.862.880.0%200KClosed weights$1.1$4.477.2%N/AN/A86.9%N/A97.9%N/A93.9%N/A87.3%
50ACQwen3.5 122B A10BAlibaba Cloud / Qwen TeamQwen3.5-122B-A10B61.457.468.656.355.672.8100.0%262.1KOpen weightsN/AN/A86.6%86.7%N/AN/AN/AN/AN/A93.4%63.8%N/A
51MELongCat Flash ThinkingMeituanLongCat-Flash-Thinking-260161.263.855.08.393.193.4100.0%128KClosed weightsN/AN/A80.5%N/A99.6%N/A82.8%N/AN/AN/A56.6%N/A
52NVNemotron Nano 9BNVIDIANemotron Nano 9B v261.234.5N/A87.175.052.980.0%131.1KOpen weightsN/AN/A64.0%N/A72.1%N/A71.1%N/AN/A90.3%N/AN/A
53MAKimi K2 ThinkingMoonshot AIKimi K2-Thinking-090559.475.440.149.567.960.2100.0%262.1KClosed weightsN/AN/A84.5%84.6%100.0%N/AN/AN/AN/AN/A60.2%N/A
54NVNemotron 3 SuperNVIDIANemotron 3 Super (120B A12B)59.170.37.078.169.470.7100.0%262.1KOpen weightsN/AN/A82.7%83.7%90.2%N/A81.2%N/AN/AN/A31.3%N/A
55ANClaude Opus 3AnthropicClaude 3 Opus58.859.3N/A60.546.272.780.0%200KClosed weightsN/AN/A50.4%68.5%N/A86.8%N/A60.1%84.9%N/AN/AN/A
56OPGPT-4.5OpenAIGPT-4.558.844.0N/A93.646.753.080.0%128KClosed weightsN/AN/A69.5%N/AN/A90.8%N/AN/A88.0%88.2%N/A36.7%
57ACQwen3 VL 235B A22B ThinkingAlibaba Cloud / Qwen TeamQwen3 VL 235B A22B Thinking58.5N/AN/A50.057.870.660.0%262.1KClosed weightsN/AN/AN/A83.8%89.7%90.6%N/AN/AN/A88.2%N/AN/A
58ACQwen3.5 27BAlibaba Cloud / Qwen TeamQwen3.5-27B57.166.837.359.452.969.9100.0%262.1KOpen weightsN/AN/A85.5%86.1%N/AN/AN/AN/AN/A95.0%61.0%N/A
59ACQwen2.5 32BAlibaba Cloud / Qwen TeamQwen2.5 32B Instruct56.930.0N/A81.768.353.680.0%N/AClosed weights$0.70$2.849.5%69.0%N/A83.3%N/A83.1%88.4%N/AN/AN/A
60OPGPT-OSS-120BOpenAIGPT OSS 120B High56.964.4N/AN/A47.956.460.0%131.1KClosed weightsN/AN/A80.9%80.7%92.5%N/AN/AN/AN/AN/AN/AN/A
61OPGPT-5.4-miniOpenAIGPT-5.4 mini56.584.9N/AN/A18.560.060.0%400KClosed weights$0.75$4.588.0%N/AN/AN/AN/AN/AN/AN/AN/AN/A
62MIMiniMax M1 80KMiniMaxMiniMax M1 80K56.345.3N/A69.469.439.980.0%1MClosed weightsN/AN/A70.0%81.1%76.9%N/A65.0%N/AN/AN/AN/A86.0%
63XIMiMo V2 FlashXiaomiMiMo-V2-Flash55.972.836.834.456.181.6100.0%262.1KOpen weights$0.14$0.2883.7%84.9%94.1%N/AN/AN/AN/AN/A58.3%N/A
64OPo3OpenAIo355.872.024.625.090.565.1100.0%200KClosed weights$2$883.3%N/A86.4%N/AN/AN/AN/AN/A49.7%91.6%
65ZAGLM 4.5 AirZhipu AIGLM-4.5-Air55.154.51.890.359.274.9100.0%131.1KOpen weightsN/AN/A75.0%81.4%N/AN/A70.7%N/AN/AN/A21.3%89.4%
66AMNova 2 LiteAmazonNova 2 Lite55.162.3N/AN/A36.867.560.0%1MClosed weights$0.33$2.7579.6%80.9%91.0%N/A71.0%N/AN/AN/AN/AN/A
67OPo4 miniOpenAIo4-mini55.067.029.8N/A69.049.780.0%200KClosed weights$1.1$4.481.4%N/A92.7%N/AN/AN/AN/AN/A51.5%93.4%
68MIMiniMax M2.1MiniMaxMiniMax M2.154.865.543.9N/A49.159.480.0%204.8KOpen weights$0.30$1.281.0%88.0%81.0%N/A78.0%N/AN/AN/A62.0%N/A
69MELlama 3.3 70BMetaLlama 3.3 70B Instruct54.625.0N/A68.666.968.780.0%128KOpen weightsN/AN/A50.5%68.9%N/A86.0%N/A77.0%88.4%92.1%N/AN/A
70ACQwen3 Next 80B A3B ThinkingAlibaba Cloud / Qwen TeamQwen3-Next-80B-A3B-Thinking54.158.4N/A37.559.562.180.0%131.1KOpen weights$0.144$1.4377.2%82.7%87.8%N/AN/AN/AN/A88.9%N/AN/A
71ACQwen3.5 35B A3BAlibaba Cloud / Qwen TeamQwen3.5-35B-A3B53.348.362.343.857.157.0100.0%262.1KOpen weightsN/AN/A84.2%85.3%N/AN/AN/AN/AN/A91.9%61.0%N/A
72ACQwen2.5 72BAlibaba Cloud / Qwen TeamQwen2.5 72B Instruct53.122.6N/A80.160.358.180.0%131.1KClosed weights$1.4$5.649.0%71.1%N/AN/A55.5%83.1%86.6%84.1%N/AN/A
73GOGemini 2.0 FlashGoogleGemini 2.0 Flash53.133.8N/A97.127.860.080.0%1MClosed weights$0.10$0.4062.1%76.4%N/AN/A35.1%89.7%N/AN/AN/AN/A
74SASarvam 105BSarvam AISarvam-105B52.860.322.867.150.065.1100.0%131.1KOpen weightsN/AN/A78.7%81.7%96.7%90.6%N/AN/AN/A84.8%49.5%N/A
75MELongCat Flash ChatMeituanLongCat-Flash-Chat52.850.4N/A54.853.153.580.0%128KClosed weightsN/AN/A73.2%82.7%61.3%89.7%48.0%N/A88.4%89.6%N/AN/A
76GOGemini 2.0 Flash LiteGoogleGemini 2.0 Flash-Lite52.626.7N/A86.0N/A51.460.0%1MClosed weights$0.075$0.3051.5%71.6%N/AN/AN/A86.8%N/AN/AN/AN/A
77MELlama 3.1 70BMetaLlama 3.1 70B Instruct52.451.8N/A55.261.237.980.0%128KClosed weightsN/AN/A41.7%66.4%N/A83.6%N/AN/A80.5%87.5%N/AN/A
78ACQwen3 VL 32B ThinkingAlibaba Cloud / Qwen TeamQwen3 VL 32B Thinking51.749.6N/AN/A47.461.160.0%N/AClosed weightsN/AN/A73.1%82.1%83.7%88.7%N/AN/AN/A87.8%N/AN/A
79ACQwen2.5 VL 32BAlibaba Cloud / Qwen TeamQwen2.5 VL 32B Instruct51.519.4N/A75.782.232.080.0%N/AClosed weightsN/AN/A46.0%68.8%N/A78.4%N/A82.2%91.5%N/AN/AN/A
80MAKimi K2Moonshot AIKimi K2-Instruct-090551.455.4N/A46.654.047.980.0%N/AClosed weightsN/AN/A75.1%81.1%49.5%89.5%53.7%N/AN/A89.8%N/A69.6%
81OPGPT-4-TurboOpenAIGPT-4 Turbo51.121.6N/A70.255.469.280.0%128KClosed weights$10$3048.0%N/AN/A86.5%N/A72.6%87.1%N/AN/AN/A
82INMercury 2InceptionMercury 250.851.9N/AN/A44.357.660.0%128KClosed weights$0.25$0.7574.0%N/A91.1%N/A67.0%N/AN/AN/AN/AN/A
83MIPhi 4 Reasoning PlusMicrosoftPhi 4 Reasoning Plus50.342.9N/AN/A66.341.560.0%N/AClosed weights$0.125$0.5068.9%76.0%78.0%N/A53.1%N/AN/A84.9%N/A81.3%
84AMNova LiteAmazonNova Lite49.246.7N/A63.134.254.680.0%300KClosed weights$0.06$0.2442.0%N/AN/A80.5%N/A73.3%85.4%89.7%N/AN/A
85MAMistral Small 4Mistral AIMistral Small 449.147.4N/AN/A61.136.160.0%256KOpen weights$0.15$0.6071.2%78.0%83.8%N/A63.6%N/AN/AN/AN/AN/A
86ACQwen3 235B A22BAlibaba Cloud / Qwen TeamQwen3-235B-A22B-Instruct-250749.059.1N/A20.852.365.780.0%262.1KOpen weightsN/AN/A77.5%83.0%70.3%N/AN/AN/AN/A88.7%N/AN/A
87GOGemini 1.5 ProGoogleGemini 1.5 Pro48.631.5N/A58.841.572.980.0%2.1MClosed weightsN/AN/A59.1%75.8%N/A85.9%N/A86.5%84.1%N/AN/AN/A
88MELlama 4 MaverickMetaLlama 4 Maverick48.544.4N/A35.743.279.580.0%1MClosed weightsN/AN/A69.8%80.5%N/A85.5%43.4%61.2%N/AN/AN/AN/A
89MIMiniMax M1 40KMiniMaxMiniMax M1 40K48.543.5N/A51.659.737.680.0%N/AClosed weightsN/AN/A69.2%80.6%74.6%N/A62.3%N/AN/AN/AN/A83.3%
90ACQwen3.6 35B A3BAlibaba Cloud / Qwen TeamQwen3.6-35B-A3B48.262.3N/A34.930.366.180.0%262.1KOpen weightsN/AN/A86.0%85.2%N/AN/AN/AN/AN/AN/AN/AN/A
91GOGemma 4 12BGoogleGemma 4 12B48.160.8N/AN/A29.352.260.0%N/AClosed weightsN/AN/A78.8%77.2%N/AN/AN/AN/AN/AN/AN/AN/A
92MIPhi 4 ReasoningMicrosoftPhi 4 Reasoning47.937.7N/AN/A73.630.560.0%N/AClosed weights$0.125$0.5065.8%74.3%62.9%N/A53.8%N/AN/A83.4%N/A75.3%
93GOGemma 2 27BGoogleGemma 2 27B47.145.594.125.68.667.0100.0%N/AClosed weightsN/AN/AN/AN/AN/A75.2%N/A42.3%51.8%N/AN/AN/A
94SASarvam 30BSarvam AISarvam-30B47.039.28.845.777.871.8100.0%128KOpen weightsN/AN/A66.5%80.0%96.7%85.1%N/AN/A92.1%N/A35.5%N/A
95ACQwen2.5 Coder 32BAlibaba Cloud / Qwen TeamQwen2.5-Coder 32B Instruct45.439.4N/A39.368.033.580.0%128KClosed weights$0.287$0.861N/A50.4%N/A75.1%31.4%57.2%92.7%N/AN/AN/A
96GOGemma 3 27BGoogleGemma 3 27B45.115.3N/A89.039.244.380.0%131.1KClosed weightsN/AN/A42.4%67.5%N/AN/A29.7%89.0%87.8%90.4%N/AN/A
97NVNemotron 3 NanoNVIDIANemotron 3 Nano (30B A3B)45.154.5N/AN/A18.365.260.0%262.1KOpen weightsN/AN/A75.0%78.3%99.2%N/AN/AN/AN/AN/AN/AN/A
98IBGranite 3.3 8BIBMGranite 3.3 8B Instruct45.0N/AN/A13.480.839.660.0%128KClosed weightsN/AN/AN/AN/AN/A65.5%N/AN/A89.7%74.8%N/A81.2%
99MIMiniMax M2MiniMaxMiniMax M244.959.715.4N/A50.252.480.0%196.6KOpen weights$0.30$1.278.0%82.0%78.0%N/A83.0%N/AN/AN/A44.0%N/A
100NVLlama 3.1 Nemotron Nano 8BNVIDIALlama 3.1 Nemotron Nano 8B V144.228.9N/A45.284.414.780.0%N/AClosed weightsN/AN/A54.1%N/A47.1%N/AN/AN/AN/A79.3%N/AN/A
101DEDeepSeek-R1DeepSeekDeepSeek-R1-052844.165.50.025.055.877.2100.0%131.1KClosed weightsN/AN/A81.0%85.0%87.5%N/A73.3%N/AN/AN/A8.9%91.4%
102ACQwen2.5 14BAlibaba Cloud / Qwen TeamQwen2.5 14B Instruct43.921.4N/A73.151.232.780.0%N/AClosed weights$0.35$1.445.5%63.7%N/A79.7%N/A80.0%83.5%N/AN/AN/A
103NVLlama 3.1 Nemotron 70BNVIDIALlama 3.1 Nemotron 70B Instruct43.833.3N/A53.2N/A48.560.0%128KOpen weightsN/AN/AN/AN/AN/A80.2%N/AN/AN/AN/AN/AN/A
104MAMistral Small 3.1 24BMistral AIMistral Small 3.1 24B Instruct43.519.079.440.759.219.6100.0%N/AClosed weightsN/AN/A46.0%66.8%N/A80.6%N/A69.3%88.4%N/AN/AN/A
105OPGPT-5.4-nanoOpenAIGPT-5.4 nano42.870.9N/AN/A8.441.760.0%400KClosed weights$0.20$1.2582.8%N/AN/AN/AN/AN/AN/AN/AN/AN/A
106ACQwen3 VL 235B A22BAlibaba Cloud / Qwen TeamQwen3 VL 235B A22B Instruct42.7N/AN/A25.044.763.760.0%262.1KClosed weightsN/AN/AN/A81.8%74.7%88.8%N/AN/AN/A87.8%N/AN/A
107MIPhi 3.5 MoEMicrosoftPhi-3.5-MoE-instruct42.441.3N/A34.339.758.480.0%N/AClosed weightsN/AN/A36.8%45.3%N/A78.9%N/A59.5%70.7%N/AN/AN/A
108ACQwen3 Next 80B A3BAlibaba Cloud / Qwen TeamQwen3-Next-80B-A3B-Instruct41.848.7N/A16.749.354.080.0%131.1KOpen weights$0.144$0.57472.9%80.6%69.5%N/AN/AN/AN/A87.6%N/AN/A
109OPGPT-4o-miniOpenAIGPT-4o mini41.512.1N/A52.556.955.380.0%128KClosed weights$0.15$0.6040.2%N/AN/A82.0%N/A70.2%87.2%N/AN/AN/A
110ANClaude Haiku 3.5AnthropicClaude 3.5 Haiku41.213.8N/A56.862.338.080.0%200KClosed weightsN/AN/A41.6%65.0%N/AN/AN/A69.4%88.1%N/AN/AN/A
111ACQwen3 VL 30B A3B ThinkingAlibaba Cloud / Qwen TeamQwen3 VL 30B A3B Thinking40.353.4N/A33.332.138.880.0%262.1KClosed weightsN/AN/A74.4%80.5%83.1%87.6%N/AN/AN/A81.7%N/AN/A
112ACQwen3 MaxAlibaba Cloud / Qwen TeamQwen3 Max40.333.2N/AN/A36.157.660.0%262.1KClosed weights$0.861$3.4462.0%N/A81.6%N/AN/AN/AN/AN/AN/AN/A
113NRHermes 3 70BNous ResearchHermes 3 70B40.229.8N/A34.0N/A65.860.0%131.1KClosed weightsN/AN/A66.1%47.2%N/A79.1%N/A20.8%N/AN/AN/AN/A
114ANClaude Sonnet 3AnthropicClaude 3 Sonnet40.048.8N/A38.720.053.680.0%200KClosed weightsN/AN/A40.4%56.8%N/A79.0%N/A43.1%73.0%N/AN/AN/A
115ACQwen3 30B A3BAlibaba Cloud / Qwen TeamQwen3 30B A3B39.737.7N/AN/A49.230.460.0%128KClosed weightsN/AN/A65.8%N/A70.9%N/A62.6%N/AN/AN/AN/A80.4%
116GOGemma 3 12BGoogleGemma 3 12B39.213.4N/A74.832.443.680.0%131.1KClosed weightsN/AN/A40.9%60.6%N/AN/A24.6%83.8%85.4%88.9%N/AN/A
117ALJamba 1.5 LargeAI21 LabsJamba 1.5 Large38.346.1N/A31.9N/A33.960.0%256KClosed weightsN/AN/A36.9%53.5%N/A81.2%N/AN/AN/AN/AN/AN/A
118GODiffusionGemma 26B A4BGoogleDiffusionGemma 26B-A4B38.350.0N/AN/A18.644.860.0%N/AClosed weightsN/AN/A73.2%77.6%N/AN/AN/AN/AN/AN/AN/AN/A
119DEDeepSeek-V3DeepSeekDeepSeek-V3 032438.142.2N/A32.336.141.980.0%163.8KClosed weightsN/AN/A68.4%81.2%N/AN/A49.2%N/AN/AN/AN/A59.4%
120ACQwen2 72BAlibaba Cloud / Qwen TeamQwen2 72B Instruct38.022.8N/A41.446.647.380.0%N/AClosed weightsN/AN/A42.4%64.4%N/A82.3%N/A59.7%86.0%N/AN/AN/A
121MIPhi 4MicrosoftPhi 436.929.7N/A56.138.521.380.0%16KClosed weights$0.125$0.5056.1%70.4%N/A84.8%N/A80.4%82.6%63.0%N/AN/A
122GOGemini 1.5 FlashGoogleGemini 1.5 Flash36.926.1N/A52.423.152.780.0%1MClosed weightsN/AN/A51.0%67.3%N/A78.9%N/A77.9%74.3%N/AN/AN/A
123AMNova MicroAmazonNova Micro36.340.7N/A51.017.334.780.0%128KClosed weights$0.035$0.1440.0%N/AN/A77.6%N/A69.3%81.1%87.2%N/AN/A
124ACQwen3.5 9BAlibaba Cloud / Qwen TeamQwen3.5-9B36.139.6N/A28.133.244.980.0%262.1KOpen weightsN/AN/A81.7%82.5%N/AN/AN/AN/AN/A91.5%N/AN/A
125ACQwen3 VL 32BAlibaba Cloud / Qwen TeamQwen3 VL 32B Instruct36.042.9N/AN/A27.036.660.0%N/AClosed weightsN/AN/A68.9%78.6%66.2%86.4%N/AN/AN/A84.7%N/AN/A
126DEDeepSeek-V3.1DeepSeekDeepSeek-V3.136.053.915.13.148.261.6100.0%131.1KOpen weightsN/AN/A74.9%83.7%49.8%N/A56.4%N/AN/AN/A30.0%66.3%
127COCommand R+CohereCommand R+35.842.4N/A10.6N/A58.260.0%128KClosed weights$0.15$0.60N/AN/AN/A75.7%N/AN/AN/AN/AN/AN/A
128MAMistral Small 3 24BMistral AIMistral Small 3 24B Instruct34.318.1N/A48.643.829.380.0%32KClosed weightsN/AN/A45.3%66.3%N/AN/AN/A70.6%84.8%82.9%N/AN/A
129ANClaude Haiku 3AnthropicClaude 3 Haiku34.137.0N/A29.227.744.580.0%200KClosed weightsN/AN/A33.3%N/AN/A75.2%N/A38.9%75.9%N/AN/AN/A
130ACQwen3 VL 8B ThinkingAlibaba Cloud / Qwen TeamQwen3 VL 8B Thinking32.944.8N/A29.214.442.580.0%262.1KClosed weightsN/AN/A69.9%77.3%80.3%85.2%N/AN/AN/A83.2%N/AN/A
131ACQwQ 32BAlibaba Cloud / Qwen TeamQwQ-32B32.636.6N/A21.034.938.380.0%N/AClosed weights$0.287$0.86165.2%N/AN/AN/A63.4%N/AN/A83.9%N/A79.5%
132ACQwen2.5 7BAlibaba Cloud / Qwen TeamQwen2.5 7B Instruct32.49.5N/A58.437.629.080.0%131.1KClosed weights$0.175$0.7036.4%56.3%N/AN/A28.7%75.5%84.8%71.2%N/AN/A
133MELlama 4 ScoutMetaLlama 4 Scout32.330.2N/A22.923.860.080.0%10MClosed weightsN/AN/A57.2%74.3%N/A79.6%32.8%50.3%N/AN/AN/AN/A
134MIPhi 4 MiniMicrosoftPhi 4 Mini31.530.1N/A35.6N/A28.360.0%128KOpen weights$0.075$0.3025.2%52.8%N/A67.3%N/A64.0%N/AN/AN/AN/A
135GOGemini 2.5 Flash LiteGoogleGemini 2.5 Flash-Lite30.735.3N/AN/A12.846.960.0%1MClosed weights$0.10$0.4064.6%N/A49.8%N/A33.7%N/AN/AN/AN/AN/A
136GOGemma 2 9BGoogleGemma 2 9B30.227.358.818.42.349.5100.0%N/AClosed weightsN/AN/AN/AN/AN/A71.3%N/A36.6%40.2%N/AN/AN/A
137MELlama 3.1 8BMetaLlama 3.1 8B Instruct29.529.3N/A20.749.215.280.0%131.1KClosed weightsN/AN/A30.4%48.3%N/A69.4%N/AN/A72.6%80.4%N/AN/A
138OPGPT-4OpenAIGPT-428.58.6N/A36.310.874.980.0%8.2KClosed weights$30$6035.7%N/AN/A86.4%N/A42.0%67.0%N/AN/AN/A
139OPGPT-4.1OpenAIGPT-4.128.438.8N/A0.038.935.180.0%1MClosed weights$2$866.3%N/A46.4%90.2%N/AN/AN/A87.4%N/A48.1%
140ACQwen3.5 4BAlibaba Cloud / Qwen TeamQwen3.5-4B28.028.9N/A18.831.334.580.0%N/AClosed weightsN/AN/A76.2%79.1%N/AN/AN/AN/AN/A89.8%N/AN/A
141OPGPT-4oOpenAIGPT-4o27.945.7N/AN/A10.321.660.0%128KClosed weightsN/AN/A70.1%74.7%N/A85.7%N/AN/AN/A81.0%N/A13.1%
142GOGemini DiffusionGoogleGemini Diffusion27.212.7N/AN/A48.423.260.0%N/AClosed weightsN/AN/A40.4%N/A23.3%N/A30.9%N/A89.6%N/AN/AN/A
143GOGemma 4 E4BGoogleGemma 4 E4B26.830.6N/AN/A18.231.860.0%131.1KOpen weightsN/AN/A58.6%69.4%N/AN/AN/AN/AN/AN/AN/AN/A
144MIPhi 3.5 miniMicrosoftPhi-3.5-mini-instruct25.932.4N/A24.418.726.980.0%128KClosed weightsN/AN/A30.4%47.4%N/A69.0%N/A48.5%62.8%N/AN/AN/A
145ACQwen3 VL 30B A3BAlibaba Cloud / Qwen TeamQwen3 VL 30B A3B Instruct25.046.1N/A8.310.830.980.0%262.1KClosed weightsN/AN/A70.4%77.8%69.3%85.0%N/AN/AN/A85.8%N/AN/A
146ACQwen2.5 Coder 7BAlibaba Cloud / Qwen TeamQwen2.5-Coder 7B Instruct24.412.1N/A19.950.516.080.0%N/AClosed weights$0.144$0.287N/A40.1%N/A67.6%18.2%46.6%88.4%N/AN/AN/A
147ACQwen3 VL 4B ThinkingAlibaba Cloud / Qwen TeamQwen3 VL 4B Thinking23.834.9N/A12.517.828.580.0%262.1KClosed weightsN/AN/A64.1%73.6%74.5%81.5%N/AN/AN/A82.6%N/AN/A
148MAMinistral 8BMistral AIMinistral 8B Instruct23.448.517.619.10.026.1100.0%131.1KOpen weightsN/AN/AN/AN/AN/A65.0%N/A54.5%34.8%N/AN/AN/A
149OPGPT-4.1-miniOpenAIGPT-4.1 mini22.535.8N/A6.318.327.980.0%1MClosed weights$0.40$1.665.0%N/A40.2%87.5%N/AN/AN/A84.1%N/A49.6%
150GOGemma 3n E4B LiteRTGoogleGemma 3n E4B Instructed LiteRT Preview22.58.932.431.716.927.1100.0%N/AClosed weightsN/AN/A23.7%50.6%11.6%64.9%13.2%N/A75.0%N/AN/AN/A
151GOGemma 3 4BGoogleGemma 3 4B22.35.2N/A51.810.826.880.0%131.1KClosed weightsN/AN/A30.8%43.6%N/AN/A12.6%75.6%71.3%90.2%N/AN/A
152ALJamba 1.5 MiniAI21 LabsJamba 1.5 Mini22.334.8N/A14.9N/A11.260.0%256.1KClosed weightsN/AN/A32.3%42.5%N/A69.7%N/AN/AN/AN/AN/AN/A
153IBIBM Granite 4.0 TinyIBMIBM Granite 4.0 Tiny Preview21.6N/AN/A7.736.919.760.0%N/AClosed weightsN/AN/AN/AN/AN/A60.4%N/AN/A82.4%63.0%N/AN/A
154ACQwen2.5 Omni 7BAlibaba Cloud / Qwen TeamQwen2.5-Omni-7B20.75.2N/A44.326.17.880.0%N/AClosed weights$0.10$0.4030.8%47.0%N/AN/AN/A71.5%78.7%N/AN/AN/A
155MELlama 3.2 3BMetaLlama 3.2 3B Instruct20.429.1N/A16.7N/A10.760.0%128KClosed weightsN/AN/A32.8%N/AN/A63.4%N/A48.0%N/A77.4%N/AN/A
156XAGrok 1.5xAIGrok-1.519.39.1N/A34.521.513.380.0%N/AClosed weightsN/AN/A35.9%51.0%N/A81.3%N/A50.6%74.1%N/AN/AN/A
157GOGemma 3n E4BGoogleGemma 3n E4B Instructed17.91.1N/A37.516.921.080.0%32KClosed weightsN/AN/A23.7%50.6%11.6%64.9%13.2%N/A75.0%N/AN/AN/A
158ACQwen3.5 2BAlibaba Cloud / Qwen TeamQwen3.5-2B16.827.2N/AN/A7.711.660.0%N/AClosed weightsN/AN/A51.6%66.5%N/AN/AN/AN/AN/A78.6%N/AN/A
159MAPixtral 12BMistral AIPixtral-12B15.2N/AN/A18.616.98.360.0%128KOpen weights$0.15$0.15N/AN/AN/A69.2%N/A48.1%72.0%61.3%N/AN/A
160ACQwen2 7BAlibaba Cloud / Qwen TeamQwen2 7B Instruct15.13.0N/A22.417.123.180.0%N/AClosed weightsN/AN/A25.3%44.1%N/A70.5%26.6%49.6%79.9%N/AN/AN/A
161GOGemma 4 E2BGoogleGemma 4 E2B14.315.9N/AN/A8.019.960.0%131.1KOpen weightsN/AN/A43.4%60.0%N/AN/AN/AN/AN/AN/AN/AN/A
162ACQwen3 VL 8BAlibaba Cloud / Qwen TeamQwen3 VL 8B Instruct12.6N/AN/A4.28.928.960.0%262.1KClosed weightsN/AN/AN/A71.6%45.9%80.7%N/AN/AN/A83.7%N/AN/A
163OPGPT-3.5-TurboOpenAIGPT-3.5 Turbo12.55.2N/A20.912.313.680.0%16.4KClosed weights$0.50$1.530.8%N/AN/A69.8%N/A43.1%68.0%N/AN/AN/A
164OPGPT-4.1-nanoOpenAIGPT-4.1 nano11.424.1N/AN/A0.05.360.0%1MClosed weights$0.10$0.4050.3%N/AN/A80.1%N/AN/AN/A74.5%N/A29.4%
165GOGemma 3n E2B LiteRTGoogleGemma 3n E2B Instructed LiteRT (Preview)8.34.88.812.37.010.1100.0%N/AClosed weightsN/AN/A24.8%40.5%6.7%60.1%13.2%N/A66.5%N/AN/AN/A
166BAERNIE 4.5BaiduERNIE 4.58.126.0N/A0.00.00.180.0%128KClosed weightsN/AN/A74.0%16.0%N/A41.9%N/A12.4%N/AN/AN/AN/A
167ACQwen3 VL 4BAlibaba Cloud / Qwen TeamQwen3 VL 4B Instruct7.7N/AN/A0.03.723.460.0%262.1KClosed weightsN/AN/AN/A67.1%46.6%77.2%N/AN/AN/A82.3%N/AN/A
168GOGemma 3n E2BGoogleGemma 3n E2B Instructed6.91.9N/A12.57.07.880.0%N/AClosed weightsN/AN/A24.8%40.5%6.7%60.1%13.2%N/A66.5%N/AN/AN/A
169GOGemma 3 1BGoogleGemma 3 1B2.70.4N/A6.91.03.280.0%N/AClosed weightsN/AN/A19.2%14.7%N/AN/A1.9%48.0%41.5%80.2%N/AN/A
170ACQwen3.5 0.8BAlibaba Cloud / Qwen TeamQwen3.5-0.8B0.30.0N/AN/A0.01.360.0%N/AClosed weightsN/AN/A11.9%42.3%N/AN/AN/AN/AN/A44.0%N/AN/A

The backend applies this page's data rule before ranking. Official PAYG prices appear here; third-party offers remain on the dedicated pricing and model detail pages.

Leaderboard highlights

Current leaders among the 170 models returned by this page's rule.

Top overall modelGPT-5.6-Sol96.5 overall scoreBest open-weight modelKimi K2.6Rank #7Lowest official inputNova Micro$0.04 / 1MLargest contextLlama 4 Scout10M tokens

About this leaderboard

Common questions about LLM Leaderboard.

What is the best model on LLM Leaderboard?

GPT-5.6-Sol is currently ranked first with a overall score of 96.5.

How is this leaderboard ranked?

The backend applies the published page rule, selects one current scored version per unique model, and returns the final rank. Benchmark cells preserve their source-native values.

How many models are included?

This page currently ranks 170 unique model products.

Where does pricing come from?

Main price columns use the model vendor's official standard PAYG API rate. Eligible third-party offers appear only in separately labeled columns, and unavailable official prices display as N/A.

Does Arena affect the score?

No. Arena results are displayed as an independent signal and are not included in the current LLMBoard capability score.

Rankings

OverallCodingText ArenaPricing

Modalities

Image GenerationVideo GenerationSpeech-to-TextEmbeddings

Benchmarks

All BenchmarksReasoningMathCoding

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai