GDPval (OpenAI economically-valuable-tasks benchmark, 0-100 win-rate scale) — 0-100 scale, 262-model corpus — distinct from Elo-based GDPval-AA family.
| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | Opus 5 | Anthropic | 67.2 | 5 | 2026-07-25 |
| 2 | GLM-5.3 | Z.ai | 63.5 | 1 | 2026-08-19 |
| 3 | Grok 4.6 | SpaceXAI | 63.3 | 4 | 2026-08-22 |
| 4 | Claude Fable 5 | Anthropic | 61.9 | 1 | 2026-06-10 |
| 5 | Qwen3.8 Max Preview | Alibaba | 61.8 | 1 | 2026-08-06 |
| 6 | GPT-5.6 Sol | OpenAI | 61.1 | 6 | 2026-07-10 |
| 7 | Qwen3.8 2.4T A95B | Alibaba | 61 | 1 | 2026-08-14 |
| 8 | Kimi K3 | Moonshot | 59.4 | 3 | 2026-07-31 |
| 9 | Claude Opus 4.7 | Anthropic | 58.6 | 2 | 2026-06-01 |
| 10 | Gemini 3.5 Flash | 57.8 | 3 | 2026-06-01 | |
| 11 | Muse Spark 1.2 | Meta | 56.4 | 1 | 2026-08-06 |
| 12 | Claude Opus 4.6 | Anthropic | 55.9 | 2 | 2026-06-01 |
| 13 | Claude Sonnet 5 | Anthropic | 54.8 | 6 | 2026-07-02 |
| 14 | Claude Sonnet 4.6 | Anthropic | 54.8 | 3 | 2026-05-02 |
| 15 | GLM-5.1-744B-A40B | Z.ai | 54.7 | 1 | 2026-06-06 |
| 16 | Claude Opus 4.8 | Anthropic | 54.2 | 1 | 2026-06-01 |
| 17 | GPT-5.6 Luna | OpenAI | 53.9 | 6 | 2026-07-10 |
| 18 | GPT-5.6 Terra | OpenAI | 53.8 | 6 | 2026-07-10 |
| 19 | DeepSeek-V4-Flash | DeepSeek | 52.9 | 5 | 2026-08-01 |
| 20 | Qwen3.8 27B | Alibaba | 52.3 | 4 | 2026-08-22 |
| 21 | Gemini 3.7 Flash | 51.6 | 3 | 2026-08-14 | |
| 22 | Grok4.5 | SpaceXAI | 51.2 | 1 | 2026-07-09 |
| 23 | Kimi K2.6 1T A32B | Moonshot | 50.4 | 1 | 2026-06-06 |
| 24 | GLM-5.2 Full Open Source | Z.ai | 50.3 | 2 | 2026-07-11 |
| 25 | Qwen3.6 Max Preview | Alibaba | 50.2 | 1 | 2026-05-02 |
| 26 | GPT-5.4 | OpenAI | 50.1 | 3 | 2026-05-02 |
| 27 | GLM-5-Turbo | Z.ai | 49.6 | 1 | 2026-05-02 |
| 28 | GPT-5.5 | OpenAI | 49.5 | 5 | 2026-06-01 |
| 29 | GLM-5.1 | Z.ai | 49.5 | 2 | 2026-05-02 |
| 30 | DeepSeek-V4-Pro | DeepSeek | 49 | 3 | 2026-06-29 |
| 31 | GPT-5.3 Codex | OpenAI | 49 | 1 | 2026-05-02 |
| 32 | GPT-5.2 | OpenAI | 48.3 | 3 | 2026-05-02 |
| 33 | MiniMax 2.7 230B A10B | MiniMax | 47.6 | 1 | 2026-06-06 |
| 34 | Claude Opus 4.5 | Anthropic | 47.3 | 2 | 2026-05-06 |
| 35 | Nemotron 3 Ultra 550B A55B | NVIDIA | 46.7 | 2 | 2026-06-06 |
| 36 | nvidia-nemotron-3-ultra-550b-a55b | NVIDIA | 46.7 | 1 | 2026-08-24 |
| 37 | Gemini 3.6 Flash | 46.1 | 1 | 2026-07-22 | |
| 38 | MiMo V2 Pro | Xiaomi | 45.2 | 1 | 2026-05-02 |
| 39 | MiniMax M3 | MiniMax | 44.8 | 2 | 2026-07-07 |
| 40 | GLM-5 | Z.ai | 44.6 | 2 | 2026-05-02 |
| 41 | Muse Spark 1.1 | Meta | 43.7 | 1 | 2026-07-11 |
| 42 | MiMo V2 Omni | Xiaomi | 42.4 | 2 | 2026-05-02 |
| 43 | GLM 5V Turbo | Z.ai | 41.4 | 1 | 2026-06-01 |
| 44 | Kimi K2.6 | Moonshot | 41.4 | 3 | 2026-07-07 |
| 45 | Gemini 3 Deep Think | 41.2 | 1 | 2026-05-02 | |
| 46 | GPT-5.4 mini | OpenAI | 40.8 | 3 | 2026-06-01 |
| 47 | Grok 4.3 | SpaceXAI | 40.6 | 5 | 2026-05-23 |
| 48 | Claude Sonnet 4.5 | Anthropic | 40.4 | 2 | 2026-06-01 |
| 49 | MiMo V2.5 Pro | Xiaomi | 39.9 | 3 | 2026-07-07 |
| 50 | GPT-5.2 Codex | OpenAI | 39.4 | 1 | 2026-05-02 |
| 51 | Solar Pro4 | Upstage | 38.8 | 1 | 2026-08-13 |
| 52 | Motif-3 | Motif Technologies | 38.7 | 1 | 2026-08-13 |
| 53 | Qwen3.7 Max | Alibaba | 38.5 | 1 | 2026-05-20 |
| 54 | Inkling-Small | Thinking Machines | 38.4 | 1 | 2026-07-31 |
| 55 | G9v3-39A5B | AI9Stars | 38.3 | 1 | 2026-08-04 |
| 56 | Kimi K2.5 | Moonshot | 38.3 | 2 | 2026-06-01 |
| 57 | JT-4.1 Flash 236B A21B | China Mobile | 38.2 | 1 | 2026-07-10 |
| 58 | Motif-3-Beta | Motif Technologies | 37.8 | 1 | 2026-07-21 |
| 59 | Nex-N2-Pro | Nex AGI | 37.8 | 2 | 2026-07-07 |
| 60 | Inkling | Thinking Machines | 36.9 | 1 | 2026-07-16 |
| 61 | Hy3-preview | Tencent | 36.5 | 2 | 2026-05-02 |
| 62 | Qwen3.5 397B A17B | Alibaba | 35.8 | 2 | 2026-06-01 |
| 63 | GPT-5 Codex | OpenAI | 35.8 | 1 | 2026-05-02 |
| 64 | Grok Build 0.1 | SpaceXAI | 35.8 | 2 | 2026-07-07 |
| 65 | Gemini 3 Flash Preview | 35.2 | 2 | 2026-05-02 | |
| 66 | GPT-5.4 nano | OpenAI | 34.9 | 3 | 2026-06-01 |
| 67 | Qwen 3.5 397B 17B | Alibaba | 34.6 | 1 | 2026-06-06 |
| 68 | GPT-5.1 Codex | OpenAI | 34.6 | 1 | 2026-05-02 |
| 69 | Kimi K2.7 Code | Moonshot | 34.5 | 2 | 2026-07-07 |
| 70 | Qwen3.5 Omni Plus | Alibaba | 34.2 | 1 | 2026-05-02 |
| 71 | Gemini 3 Pro | 34.2 | 2 | 2026-05-02 | |
| 72 | MiniMax M2.5 | MiniMax | 33.9 | 1 | 2026-05-02 |
| 73 | Agnes 2.5 Pro Alpha | Sapiens AI | 33.8 | 1 | 2026-07-25 |
| 74 | GLM-4.7 | Z.ai | 33.7 | 2 | 2026-06-01 |
| 75 | Grok 4.20 0309 V2 | SpaceXAI | 33.5 | 2 | 2026-05-02 |
| 76 | MiniMax M2.7 | MiniMax | 33 | 2 | 2026-07-07 |
| 77 | Qwen3.5 27B | Alibaba | 33 | 2 | 2026-05-02 |
| 78 | GPT-5 | OpenAI | 32.5 | 4 | 2026-07-03 |
| 79 | Muse Spark | Meta | 32.3 | 2 | 2026-07-07 |
| 80 | GPT-5.5 Instant | OpenAI | 32.2 | 3 | 2026-07-07 |
| 81 | Qwen3.6 Plus | Alibaba | 32 | 2 | 2026-07-07 |
| 82 | Qwen3.6 27B | Alibaba | 32 | 2 | 2026-05-02 |
| 83 | Gemini 3.5 Flash Lite | 32 | 1 | 2026-07-22 | |
| 84 | Claude Haiku 4.5 | Anthropic | 31.8 | 2 | 2026-06-01 |
| 85 | Claude Sonnet 4 | Anthropic | 31.2 | 2 | 2026-06-01 |
| 86 | Solar Open2 250B | Upstage | 31.1 | 1 | 2026-08-13 |
| 87 | A.X-K2 | SK Telecom | 30.5 | 1 | 2026-08-13 |
| 88 | Ling-3.0-flash | InclusionAI | 30.4 | 1 | 2026-08-06 |
| 89 | MiniMax M2.1 | MiniMax | 29.7 | 1 | 2026-05-02 |
| 90 | MiMo V2 Flash | Xiaomi | 29 | 3 | 2026-05-06 |
| 91 | JT-35B-Flash | China Mobile | 28.8 | 1 | 2026-05-13 |
| 92 | DeepSeek-V3.1 | DeepSeek | 28.7 | 2 | 2026-06-01 |
| 93 | Gemini 2.5 Flash Preview 09-2025 | 28.5 | 2 | 2026-06-01 | |
| 94 | DeepSeek-V3.2-Exp | DeepSeek | 28.3 | 2 | 2026-06-01 |
| 95 | Qwen3.6 35B A3B | Alibaba | 27.8 | 2 | 2026-06-01 |
| 96 | GPT-5.1 Codex mini | OpenAI | 27.7 | 1 | 2026-05-02 |
| 97 | Step 3.5 Flash | StepFun | 27.6 | 1 | 2026-05-02 |
| 98 | Claude 3.7 Sonnet | Anthropic | 27.4 | 2 | 2026-05-02 |
| 99 | ling-2-6-1t | InclusionAI | 27.2 | 1 | 2026-05-02 |
| 100 | Grok 4.20 | SpaceXAI | 27 | 1 | 2026-05-02 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.