ModelBench
Model directory

All documented models

37 public models with source-linked specifications, provider pricing, and benchmark records.

Ranked by raw score for SWE-Bench Verified · v. Versions are kept separate.
Filterable model directory
ModelCreatorRaw benchmark scoreModalitiesContextInput / 1MOutput / 1MReleased
Claude Fable 5anthropic/claude-fable-5Anthropic95.0Image, Pdf, Text1M$3.000$18.0002026-06-09
Claude Opus 4.8anthropic/claude-opus-4-8Anthropic88.6Image, Pdf, Text1M$0.425$2.1252026-05-28
Claude Sonnet 5anthropic/claude-sonnet-5Anthropic85.2Image, Pdf, Text1M$1.440$7.2002026-06-30
Ornith 1.0 397Bdeepreinforce/ornith-1.0-397bDeepreinforce82.4Image, Text262.144K$0.000$0.0002026-06-25
DeepSeek V4 Prodeepseek/deepseek-v4-proDeepSeek80.6Text1M$0.348$0.6962026-04-24
MiniMax-M3minimax/MiniMax-M3MiniMax80.5Image, Text, Video512K$0.300$1.2002026-06-01
Qwen3.7 Maxalibaba/qwen3.7-maxAlibaba Qwen80.4Text1M$1.250$3.7502026-05-21
Kimi K2.6moonshotai/kimi-k2.6Moonshot AI80.2Image, Text, Video262.144K$0.275$1.1002026-04-21
MiniMax-M2.7minimax/MiniMax-M2.7MiniMax79.9Text204.8K$0.240$0.9602026-03-18
DeepSeek V4 Flashdeepseek/deepseek-v4-flashDeepSeek79.0Text1M$0.062$0.1252026-04-24
MiMo-V2.5-Proxiaomi/mimo-v2.5-proXiaomi78.9Text1.04858M$0.400$0.8002026-04-22
Mistral Medium (latest)mistral/mistral-medium-latestMistral AI77.6Image, Text262.144K$0.400$2.0002026-04-29
Mistral Medium 3.5mistral/mistral-medium-2604Mistral AI77.6Image, Text262.144K$1.500$7.5002026-04-29
Qwen3.6 27Balibaba/qwen3.6-27bAlibaba Qwen77.2Audio, Image, Text, Video262.144K$0.200$1.5002026-04-22
Step 3.7 Flashstepfun/step-3.7-flashStepFun76.5Image, Text256K$0.200$1.1502026-05-29
Qwen3.5 397B-A17Balibaba/qwen3.5-397b-a17bAlibaba Qwen76.4Audio, Image, Text, Video262.144K$0.172$1.0322026-02-15
MiniMax-M2.5minimax/MiniMax-M2.5MiniMax75.8Text204.8K$0.110$0.9502026-02-12
Ornith 1.0 35Bdeepreinforce/ornith-1.0-35bDeepreinforce75.6Image, Text262.144K$0.000$0.0002026-06-25
Hy3 previewtencent/hy3-previewTencent74.4Text256K$0.063$0.2102026-04-20
Step 3.5 Flashstepfun/step-3.5-flashStepFun74.4Text256K$0.090$0.3002026-01-29
MiniMax-M2.1minimax/MiniMax-M2.1MiniMax74.0Text204.8K$0.270$1.1002025-12-23
GLM-4.7zhipuai/glm-4.7Zhipu AI73.8Text204.8K$0.200$0.8002025-12-22
Qwen3.6 35B-A3Balibaba/qwen3.6-35b-a3bAlibaba Qwen73.4Audio, Image, Text, Video262.144K$0.140$1.0002026-04-17
GLM-5zhipuai/glm-5Zhipu AI72.8Text204.8K$0.480$1.9002026-02-12
Qwen3.5 27Balibaba/qwen3.5-27bAlibaba Qwen72.4Audio, Image, Text, Video262.144K$0.086$0.6882026-02-23
Qwen3.5 122B-A10Balibaba/qwen3.5-122b-a10bAlibaba Qwen72.0Audio, Image, Text, Video262.144K$0.115$0.9172026-02-23
MAI-Code-1-Flashmicrosoft/mai-code-1-flashMicrosoft71.6Text256K$0.750$4.5002026-06-02
Kimi K2 Thinkingmoonshotai/kimi-k2-thinkingMoonshot AI71.3Text262.144K$0.300$1.2002025-11-06
Laguna XS 2.1poolside/laguna-xs-2.1Poolside70.9Text262.144K$0.060$0.1202026-07-02
Kimi K2.5moonshotai/kimi-k2.5Moonshot AI70.8Image, Text, Video262.144K$0.300$1.5002026-01
Nemotron 3 Ultra 550B A55Bnvidia/nemotron-3-ultra-550b-a55bNVIDIA70.7Text1M$0.100$0.1002026-06-04
MiniMax-M2minimax/MiniMax-M2MiniMax69.4Text196.608K$0.200$1.0002025-10-27
Ornith 1.0 9Bdeepreinforce/ornith-1.0-9bDeepreinforce69.4Image, Text262.144K$0.000$0.0002026-06-25
North Mini Codecohere/north-mini-code-1-0Cohere67.6Text256K$0.000$0.0002026-06-09
Devstral Mediummistral/devstral-medium-2507Mistral AI61.6Text128K$0.400$2.0002025-07-10
GLM-4.7-Flashzhipuai/glm-4.7-flashZhipu AI59.2Text200K$0.040$0.3002026-01-19
Devstral Smallmistral/devstral-small-2507Mistral AI53.6Text128K$0.100$0.3002025-07-10