Coding
PostTrainBench v1 leaderboard
As of 2026-11-18, the highest measured score on PostTrainBench v1 is 34.3% by GLM-5.2. 133 more models have estimated scores, calibrated from the benchmarks they were measured on.
Measured scores: benchlm.ai.
| # | Model | Score | Source |
|---|---|---|---|
| 1 | Claude Opus 5.5 | 47.7% | estimated ± 1.3 pp, low confidence |
| 2 | Claude Sonnet 5.5 | 47.4% | estimated ± 4.6 pp, medium confidence |
| 3 | Claude Mythos 5 | 46.4% | estimated ± 4.6 pp, medium confidence |
| 4 | Claude Fable 5.1 | 45.9% | estimated ± 1.8 pp, low confidence |
| 5 | GPT-6 Astra | 45.7% | estimated ± 1.3 pp, low confidence |
| 6 | MiniMax M3.1 Flash | 44.8% | estimated ± 1.8 pp, low confidence |
| 7 | Gemini 4 Carbon | 44.8% | estimated ± 1.3 pp, low confidence |
| 8 | Gemini 4 Argon | 44.2% | estimated ± 1.3 pp, low confidence |
| 9 | Grok 4.6 | 43.7% | estimated ± 3.6 pp, medium confidence |
| 10 | GPT-5.6 Terra | 43.4% | estimated ± 3.6 pp, medium confidence |
| 11 | Claude Fable 5 | 42.9% | estimated ± 3.6 pp, medium confidence |
| 12 | GPT-6.1 Sol | 42.5% | estimated ± 1.3 pp, low confidence |
| 13 | GPT-5.6 Luna | 40.5% | estimated ± 3.6 pp, medium confidence |
| 14 | GLM-5.3-Flash | 40.4% | estimated ± 1.3 pp, low confidence |
| 15 | Kimi K3.1 Pro | 40.1% | estimated ± 1.3 pp, low confidence |
| 16 | Sakana Fugu-Ultra | 39.5% | estimated ± 4.6 pp, medium confidence |
| 17 | Claude Opus 5 | 39.2% | estimated ± 1.3 pp, low confidence |
| 18 | Hy4 | 38.7% | estimated ± 1.3 pp, low confidence |
| 19 | GLM-5.3 | 38.0% | estimated ± 1.3 pp, low confidence |
| 20 | Ember-1 | 35.4% | estimated ± 5.5 pp, low confidence |
| 21 | Inkling 2 Small | 35.3% | estimated ± 1.3 pp, low confidence |
| 22 | DeepSeek V4 Flash 0731 | 35.3% | estimated ± 3.6 pp, medium confidence |
| 23 | Kimi K3 | 34.9% | estimated ± 1.3 pp, low confidence |
| 24 | GPT-5.6 Sol | 34.4% | estimated ± 1.3 pp, low confidence |
| 25 | Ling 3.1 Flash | 34.3% | estimated ± 4.6 pp, high confidence |
| 26 | GLM-5.2 | 34.3% | measured |
| 27 | Claude Opus 4.8 | 34.1% | measured |
| 28 | Muse Spark 1.2 | 32.5% | estimated ± 3.6 pp, high confidence |
| 29 | Qwen3.8-27B | 31.8% | estimated ± 3.6 pp, high confidence |
| 30 | Beam 1.0 501B-A23B | 31.6% | estimated ± 4.6 pp, high confidence |
| 31 | Beam 1.0 501B-A23B FP8 | 31.4% | estimated ± 4.6 pp, high confidence |
| 32 | Qwen3.8 Max | 31.3% | estimated ± 3.6 pp, high confidence |
| 33 | Hy4 preview | 31.2% | estimated ± 4.6 pp, high confidence |
| 34 | Beam 1.0 501B-A23B NVFP4 | 31.1% | estimated ± 4.6 pp, high confidence |
| 35 | Ornith-1.5-397B | 30.5% | estimated ± 4.6 pp, high confidence |
| 36 | Mistral Large 4 | 29.8% | estimated ± 3.1 pp, low confidence |
| 37 | Qwen3.8-Omni-Flash | 28.7% | estimated ± 4.6 pp, high confidence |
| 38 | Claude Opus 4.7 | 28.6% | measured |
| 39 | Qwen3.8-Flash-Next | 27.8% | estimated ± 4.6 pp, high confidence |
| 40 | Ornith-1.0-397B | 27.5% | estimated ± 4.6 pp, high confidence |
| 41 | Muse Wonder | 27.2% | estimated ± 1.3 pp, medium confidence |
| 42 | Inkling-Small | 27.1% | estimated ± 3.6 pp, high confidence |
| 43 | Muse Spark 1.1 | 26.8% | estimated ± 3.6 pp, high confidence |
| 44 | Ripple-65B Preview | 26.7% | estimated ± 4.6 pp, high confidence |
| 45 | Sage-A1.5 | 26.5% | estimated ± 4.6 pp, high confidence |
| 46 | dots3-note Preview | 26.3% | estimated ± 4.6 pp, high confidence |
| 47 | Grok 4.5 | 25.6% | estimated ± 1.3 pp, medium confidence |
| 48 | Gemini 3.7 Flash | 25.4% | estimated ± 3.6 pp, high confidence |
| 49 | GPT-5.5 | 25.0% | measured |
| 50 | Atria Dawn Preview | 24.8% | estimated ± 4.6 pp, high confidence |
| 51 | Ornith-1.5-35B-A3B | 24.8% | estimated ± 4.6 pp, high confidence |
| 52 | Laguna S 2.1 | 24.6% | estimated ± 4.6 pp, high confidence |
| 53 | Gemini 3.8 Flash | 24.4% | estimated ± 3.6 pp, high confidence |
| 54 | Sakana Fugu | 24.2% | estimated ± 4.6 pp, high confidence |
| 55 | Claude Sonnet 5 | 23.9% | estimated ± 3.6 pp, high confidence |
| 56 | Composer 2.5 | 23.9% | estimated ± 3.6 pp, high confidence |
| 57 | Gemini 3.6 Flash | 23.9% | estimated ± 3.6 pp, high confidence |
| 58 | Claude Opus 4.6 | 23.2% | measured |
| 59 | Qwen3.7 Plus | 22.7% | estimated ± 4.6 pp, high confidence |
| 60 | Kimi K2.7 Code | 22.1% | estimated ± 3.6 pp, high confidence |
| 61 | Gemini 3.1 Pro | 21.6% | measured |
| 62 | Inkling | 21.4% | estimated ± 3.6 pp, high confidence |
| 63 | GPT-5.2 | 21.4% | measured |
| 64 | Step 3.7 Flash | 21.4% | estimated ± 4.6 pp, high confidence |
| 65 | GPT-5.4 | 20.2% | measured |
| 66 | Claude Opus 4.6 (Adaptive) | 20.2% | estimated ± 3.1 pp, medium confidence |
| 67 | GLM-5 | 20.1% | estimated ± 4.6 pp, high confidence |
| 68 | Claude Sonnet 4.6 | 20.0% | estimated ± 3.1 pp, medium confidence |
| 69 | DeepSeek V4 Pro | 19.8% | estimated ± 3.1 pp, medium confidence |
| 70 | Gemini 3.5 Flash | 19.7% | estimated ± 3.1 pp, medium confidence |
| 71 | GPT-5.1-Codex-Max | 19.7% | measured |
| 72 | GPT-5.4 mini | 19.6% | estimated ± 3.1 pp, medium confidence |
| 73 | Spark-X2.6-4B | 19.3% | estimated ± 4.6 pp, high confidence |
| 74 | Kimi K2.6 | 19.1% | estimated ± 3.1 pp, medium confidence |
| 75 | GLM-5.1 | 19.0% | estimated ± 3.1 pp, medium confidence |
| 76 | MiniMax M2.7 | 19.0% | estimated ± 3.1 pp, medium confidence |
| 77 | GPT-5.4 nano | 19.0% | estimated ± 3.1 pp, medium confidence |
| 78 | Qwen3.6 Plus | 19.0% | estimated ± 3.1 pp, medium confidence |
| 79 | GPT-5.1 | 19.0% | estimated ± 3.1 pp, medium confidence |
| 80 | Claude Sonnet 4.5 Thinking | 19.0% | estimated ± 3.1 pp, medium confidence |
| 81 | Claude Opus 4.5 Thinking | 19.0% | estimated ± 3.1 pp, medium confidence |
| 82 | Gemini 3 Flash | 19.0% | estimated ± 3.1 pp, medium confidence |
| 83 | GPT-5 | 19.0% | estimated ± 3.1 pp, medium confidence |
| 84 | Muse Spark | 19.0% | estimated ± 3.1 pp, medium confidence |
| 85 | Qwen3.5 Plus | 19.0% | estimated ± 3.1 pp, medium confidence |
| 86 | MiniMax M2.5 | 19.0% | estimated ± 3.1 pp, medium confidence |
| 87 | GPT-5 mini | 19.0% | estimated ± 3.1 pp, low confidence |
| 88 | GPT-5.1-Codex | 19.0% | estimated ± 3.1 pp, low confidence |
| 89 | Claude Haiku 4.5 Thinking | 19.0% | estimated ± 3.1 pp, low confidence |
| 90 | Grok 4.20 | 19.0% | estimated ± 3.1 pp, low confidence |
| 91 | Qwen3 Max | 19.0% | estimated ± 3.1 pp, low confidence |
| 92 | GLM-4.6 | 19.0% | estimated ± 3.1 pp, low confidence |
| 93 | Grok 4.1 Fast (Reasoning) | 19.0% | estimated ± 3.1 pp, low confidence |
| 94 | Gemini 2.5 Pro | 19.0% | estimated ± 3.1 pp, low confidence |
| 95 | Gemini 3.1 Flash-Lite | 19.0% | estimated ± 3.1 pp, low confidence |
| 96 | Grok 4 Fast (Reasoning) | 19.0% | estimated ± 3.1 pp, low confidence |
| 97 | Gemini 3.5 Flash-Lite | 18.2% | estimated ± 3.6 pp, high confidence |
| 98 | MiniMax M3 | 18.2% | estimated ± 3.6 pp, high confidence |
| 99 | Gemini 3 Pro | 18.1% | measured |
| 100 | GPT-5.3 Codex | 17.8% | measured |
| 101 | MAI-Thinking-1 | 17.7% | estimated ± 4.6 pp, medium confidence |
| 102 | GPT-5.2-Codex | 17.2% | measured |
| 103 | BTL-4 | 17.2% | estimated ± 5.5 pp, low confidence |
| 104 | Claude Opus 4.5 | 17.1% | measured |
| 105 | MiMo-V2.5-Pro | 16.9% | estimated ± 3.6 pp, high confidence |
| 106 | MiMo-V2-Pro | 16.8% | estimated ± 5.5 pp, low confidence |
| 107 | Nemotron 3.5 Super VL 120B A12B | 16.7% | estimated ± 4.6 pp, medium confidence |
| 108 | Apodex 1.1 | 16.5% | estimated ± 5.5 pp, low confidence |
| 109 | Nemotron 3.5 Super VL 120B A12B NVFP4 | 16.2% | estimated ± 4.6 pp, medium confidence |
| 110 | Muse Glimmer 30B | 16.1% | estimated ± 4.6 pp, medium confidence |
| 111 | Claude Sonnet 4.5 | 16.0% | estimated ± 5.5 pp, low confidence |
| 112 | Qwen3.5 397B | 15.7% | estimated ± 4.6 pp, medium confidence |
| 113 | Kimi K2.5 | 15.5% | estimated ± 4.6 pp, medium confidence |
| 114 | Qwen 3.6 Max (preview) | 15.5% | estimated ± 3.6 pp, high confidence |
| 115 | Ornith-1.0-35B | 15.2% | estimated ± 4.6 pp, medium confidence |
| 116 | Qwen3.6-35B-A3B | 14.3% | estimated ± 4.6 pp, medium confidence |
| 117 | MiMo-V2-Omni | 13.7% | estimated ± 5.5 pp, low confidence |
| 118 | Grok 4.3 | 13.7% | estimated ± 3.6 pp, medium confidence |
| 119 | Claude 4.1 Opus | 13.5% | estimated ± 5.5 pp, low confidence |
| 120 | Hy3 Preview | 13.4% | estimated ± 5.5 pp, low confidence |
| 121 | Step 3.5 Flash | 13.4% | estimated ± 5.5 pp, low confidence |
| 122 | MiMo-V2.5 | 13.2% | estimated ± 3.6 pp, medium confidence |
| 123 | Flint | 13.2% | estimated ± 5.5 pp, low confidence |
| 124 | MiMo-V2-Flash | 12.6% | estimated ± 5.5 pp, low confidence |
| 125 | Laguna XS 2.1 | 12.3% | estimated ± 4.6 pp, medium confidence |
| 126 | Ornith-1.5-9B | 12.2% | estimated ± 4.6 pp, medium confidence |
| 127 | Jais-3-MoVA-36B-A4B | 12.1% | estimated ± 5.5 pp, low confidence |
| 128 | Claude 4 Sonnet | 12.0% | estimated ± 5.5 pp, low confidence |
| 129 | Qwen3.6-27B | 12.0% | estimated ± 3.6 pp, medium confidence |
| 130 | MAI-Code-1.1-Flash | 11.9% | estimated ± 5.5 pp, low confidence |
| 131 | Claude 4 Opus | 11.8% | estimated ± 5.5 pp, low confidence |
| 132 | Jais-3-32B | 11.8% | estimated ± 5.5 pp, low confidence |
| 133 | Qwen3.5-27B | 11.8% | estimated ± 5.5 pp, low confidence |
| 134 | Qwen3.5-122B-A10B | 11.5% | estimated ± 5.5 pp, low confidence |
| 135 | GLM-4.7 | 11.2% | estimated ± 3.6 pp, medium confidence |
| 136 | Laguna XS.2 | 11.0% | estimated ± 4.6 pp, medium confidence |
| 137 | Nemotron 3 Ultra | 10.7% | estimated ± 3.6 pp, medium confidence |
| 138 | Qwen3.7 Max | 10.5% | estimated ± 3.6 pp, medium confidence |
| 139 | Claude Haiku 4.5 | 7.8% | estimated ± 3.6 pp, medium confidence |
| 140 | Mistral Medium 3.5 128B | 7.5% | estimated ± 3.6 pp, medium confidence |
| 141 | Ornith-1.0-9B | 7.4% | estimated ± 4.6 pp, medium confidence |
| 142 | Ling 3.0 Flash | 6.0% | estimated ± 3.6 pp, medium confidence |
| 143 | Qwen3.5 Flash | 5.1% | estimated ± 3.6 pp, medium confidence |
| 144 | LongCat-Flash-Lite-Sparse | 5.1% | estimated ± 4.6 pp, medium confidence |
| 145 | Grok 4 | 0.0% | estimated ± 3.6 pp, medium confidence |
| 146 | Laguna M.1 | 0.0% | estimated ± 3.6 pp, medium confidence |
All leaderboards
Agentic · terminal
- AA Terminal-Bench 2.1
- AA Terminal-Bench 4.0
- Terminal-Bench 4.0
- Terminal-Bench 2.0
- Terminal-Bench 2.1
- Terminal-Bench 3.0
- Terminal-Bench Hard
- Terminal-Bench-Science 0.1
- Terminal-Bench 4.1
- Terminal-Bench 2.1 (Vals AI)
Agentic · tools
- AA Agentic Index
- AA-AnalystAgent
- AA AutomationBench
- CWE-Bench-AA
- AA Cyber Index
- DeepsecBench-AA
- AA EnterpriseOps-Gym
- GDP.pdf
- AA Harvey LAB v1.0
- Harvey LAB-AA v1.1
- AA ITBench
- AA Tau3 Banking
- Agents' Last Exam
- APEX-Agents
- APEX-Agents-AA
- ApprenticeBench
- AutomationBench
- BFCL v4
- BrowseComp
- Claw-Eval
- CWE-bench v1
- CWE-bench
- CyberGym
- DeepPlanning
- DeepSearchQA
- DRACO
- ExploitGym
- Finance Agent v2
- GDPval-AA
- Gert Labs
- Hermes Bench
- Hermes Index
- HLE w/ tools
- JobBench
- LifeVal
- MazeBench
- MCP Atlas
- MCP-Tasks
- MKA
- OSWorld 2.1
- OSWorld 2.1 Offline
- OSWorld-Verified
- OSWorld 2.0
- PinchBench
- QwenClawBench
- ResearchClawBench
- SkillsBench
- τ²-bench results
- τ³-bench results
- Toolathlon
- Toolathlon-Verified
- VITA-Bench
- WideResearch
Coding
- AA Coding Index
- AA LiveCodeBench
- AA-SciCode
- Bug Hunt Bench
- CursorBench 3.1
- CursorBench 3.2
- CursorBench 4.0
- DeepSWE
- FrontierCode 1.1 Main
- FrontierCode 2.0
- FrontierCode 1.1 Extended
- FrontierSWE v2
- LiveCodeBench
- LiveCodeBench Pro
- LiveCodeBench v6
- NL2Repo
- OpenHarmony Bench v1.2
- OpenHarmony Bench v1.0
- PostTrainBench
- PostTrainBench v1
- PostTrainBench v1.1
- ProgramBench
- React Native Evals
- ScarfBench
- SciCode
- SWE-Atlas Codebase QnA
- SWE-bench Pro
- SWE-bench Verified
- SWE-Rebench
- SWE-Marathon
- Vals LiveCodeBench
- Vals SWE-bench
- Vibe Code Bench
- VulcanBench Frontier v4
- VulcanBench v3
Math
- AIME 2024
- AIME 2025
- AIME26
- FrontierMath v2 (Tier 4)
- FrontierMath v2 (Tiers 1-3)
- HMMT Feb 2025
- HMMT Feb 2026
- HMMT Nov 2025
- IMOAnswerBench
- MATH-500
- MMAnswerBench
Knowledge & reasoning
- AA-GPQA Diamond
- HealthBench Professional (length-adjusted, Artificial Analysis)
- HealthBench Professional (raw, Artificial Analysis)
- AA-HLE
- AA MMLU-Pro
- ARC-AGI-2
- ARC-AGI-3 Provider Adapter
- ARC-AGI-1
- ARC-AGI-3
- Artificial Analysis Intelligence Index
- BBH
- BioBasicsBench
- BioMysteryBench (human-difficult)
- BioMysteryBench (human-solvable)
- C-Eval
- CritPt
- GPQA
- GPQA Diamond
- HealthBench (raw)
- HealthBench Hard
- HealthBench (length-adjusted)
- HealthBench Professional
- HealthBench Professional (raw)
- HLE
- HLE w/o tools
- HLE-Verified
- JevBench 1.4
- LABBench2
- MedXpertQA (Text)
- MMLU
- MMLU-Pro
- MMLU-Pro (Arcee)
- MMLU-Redux
- MMMLU
- AA-Omniscience Accuracy
- SimpleQA
- STEMBench
- SuperGPQA
- Vals GPQA Diamond
- Vals MMLU-Pro
Instruction following
Multilingual
Vision & documents
- AA-MMMU-Pro
- AI2D_TEST
- BabyVision
- BabyVision w/ Python
- BenchCAD Vision2Code (no tools)
- BenchCAD Vision2Code (tools)
- CC-OCR
- Chartography (no tools)
- Chartography (tools)
- CharXiv
- CharXiv w/o tools
- CountBench
- DynaMath
- ERQA
- LVBench
- MathVision
- MathVision w/ Python
- MedXpertQA (MM)
- MLVU (M-Avg)
- MMMU
- MMMU-Pro
- MMMU-Pro w/ Python
- MMVU
- OCRBench V2
- OfficeQA
- OfficeQA Pro
- OmniDocBench 1.5
- ParseBench-1
- ParseBench-2
- PerceptionBench
- RealWorldQA
- RefCOCO (avg)
- ScreenSpot Pro
- SimpleVQA
- SWE-bench Multimodal
- Video-MME (w/o subtitle)
- Video-MME (with subtitle)
- VideoMMMU
- V*
- We-Math
- ZeroBench
Long context
General
- HLE-MultiChoice (Decision Index)
- Artificial Analysis Decision Index v1.0
- Artificial Analysis Decision Index
- When2Call (Decision Index)
- WildGuardTest (Decision Index)
- BabaIsBench
- BinaryAudit
- BlenderWorld (text-only)
- Chess Puzzles
- CompileBench
- Cookie-Bench
- DAYJOB
- DAYJOB Healthcare
- Decision Index Full Score
- EBR-bench
- Epoch Automation Reports
- ExploitBench
- ExploitBench Internal Port
- FACTS Multimodal
- FailureSensorIQ
- FrontierPhysics
- Furniture Assembly
- Decision Index Arts
- Decision Index Knowledge
- Decision Index Language
- Decision Index Retrieval
- Decision Index Tools
- MaxxBench
- MirrorCode
- Mystery Game Puzzles
- NYT Connections
- OTelBench
- SEC-Bench Pro
- SkillsBench official leaderboard (BenchFlow)
- SpireClimb
- StealthMark