Benchmarks
357 sources · 1,263 models × 3,079 benchmarks · 162,127 scoresSurge's Chartography board as rescored between its November 6 and November 18, 2026 site builds, on the 100 professional chart-reading tasks. The page has no changelog and names no metric, protocol, judge, tool access or run count. Surge's GitHub harness, unchanged since August 14, still describes ten no-tools runs per task judged by Gemini 3.5 Flash, while a Harbor-format release of the tasks, each a container holding the chart image, appeared on Hugging Face on November 10, and the launch post still says the best model reaches 45%. Kept apart from the original board because the runs it kept moved by as much as 59.9 points (Grok 4.7 xHigh, 14.7 to 74.6; Claude Opus 5.5 Max, 66.3 to 89.6).
253rd strongest of 818 fitted benchmarks
Signal-to-noise +9.03
329th hardest of 818 fitted benchmarks
Capability needed 156.6
| # | Model | Setting | Score | Source |
|---|---|---|---|---|
| 1 | GPT-6.2 Astra | max | 90.7 | |
| 2 | Claude Opus 5.5 | max | 89.6 | |
| 3 | GPT-6 Astra | max | 88.6 | |
| 4 | GPT-6.2 Sol | max | 87.7 | |
| 5 | Gemini 4 Hydrogen | x-high | 87.2 | |
| 6 | Gemini 4 Neon | high | 86.6 | |
| 7 | Gemini 4 Carbon | high | 85.5 | |
| 8 | Muse Spark 1.3 | max | 84.6 | |
| 9 | Claude Opus 5 | max | 84.4 | |
| 10 | Gemini 3.8 Flash | high | 81.7 | |
| 11 | Claude Fable 5.2 | max | 80.8 | |
| 12 | Claude Fable 5.1 | max | 79.6 | |
| 13 | GPT-6 Sol | max | 79.4 | |
| 14 | Muse Ember | max | 79.3 | |
| 15 | Claude Fable 5 | max | 79.0 | |
| 16 | GPT-5.6 Sol | max | 78.6 | |
| 17 | Inkling 2 Small | 0.99 | 75.6 | |
| 18 | Qwen4-Max | x-high | 75.3 | |
| 19 | Gemini 3.7 Flash | high | 74.9 | |
| 20 | Grok 4.7 | x-high | 74.6 | |
| 21 | GLM-5.3 Flash | max | 72.7 | |
| 22 | GPT-5.6 Luna | max | 72.2 | |
| 23 | GPT-6 Luna | max | 68.0 | |
| 24 | Muse Spark 1.2 | x-high | 66.7 | |
| 25 | Kimi K3 | max | 66.4 | |
| 26 | GLM-5.3 | max | 64.2 | |
| 27 | Sonnet 5 | max | 64.0 | |
| 28 | GPT-5.6 Terra | x-high | 58.8 | |
| 29 | Gemini 3.1 Pro Preview | high | 54.2 | |
| 30 | Hy4 | high | 49.2 | |
| 31 | Grok 4.6 | x-high | 44.4 | |
| 32 | MiniMax M3.1 Flash | max | 39.8 | |
| 33 | Hy3 | high | 20.3 | |
| 34 | Muse Glimmer 30B | x-high | 15.4 | |
| 35 | Inkling | high | 15.0 | |
| 36 | Mistral Large 3 675B Instruct 2512 | not supported | 7.3 |
Benchmarks outside its family whose model rankings track it most reliably (Spearman ρ over at least 10 shared models).