Benchmarks · coding
LiveCodeBench (Vals)
Our Implementation of the LiveCodeBench benchmark Scores are the values published by VALS AI. Source benchmark version 1.
56 models · 56 results · updated 2026-09-01 · vals.ai ↗
Leaderboard
Fable 5.1 max
Heavy line: best score at each price. Thin lines join one model's effort levels. Cost uses each model's median subscription plan, or API price with no plan data. 38 more models under All models in settings.
More from this benchmark
4 published measuresTime per task143 results · lower is better
40 more models under All models in settings.
Score vs any published measure3 measures: sub-scores, splits, counts
Breakdown
darker is better within each column · click a column to sort| Model | |||
|---|---|---|---|
| Fable 5.1 max | 99.1 | 91.9 | 80.6 |
| Fable 5 max | 98.8 | 90.9 | 79.7 |
| Gemini 3.8 Flash high | 99.1 | 91.4 | 78.0 |
| Opus 5 max | 99.1 | 90.6 | 77.4 |
| Gemini 3.7 Flash high | 99.1 | 90.6 | 76.3 |
| Grok 4.6 high | 99.1 | 89.0 | 76.6 |
| Gemini 3.6 Flash high | 99.1 | 90.6 | 74.6 |
| GPT-5.2 Codex xhigh | 99.1 | 90.3 | 74.6 |
| Qwen 3.8 Max | 99.1 | 88.8 | 75.7 |
| Opus 4.8 max | 99.1 | 89.8 | 74.6 |
| Gemini 3.5 Flash high | 99.1 | 90.6 | 73.1 |
| DeepSeek V4 Pro 0813 max | 98.8 | 89.8 | 74.0 |
| Grok 4.5 high | 99.1 | 89.6 | 73.4 |
| GPT-5.3 Codex xhigh | 99.1 | 87.7 | 75.1 |
| DeepSeek V4 Flash 0731 high | 98.8 | 89.0 | 74.0 |
| Kimi K3 max | 98.8 | 88.5 | 74.3 |
| Qwen 3.7 Max | 98.4 | 89.3 | 73.4 |
| Kimi K2.6 max | 99.1 | 89.8 | 71.4 |
| Qwen 3.6 Plus max | 99.1 | 91.6 | 67.1 |
| Inkling Small max | 99.1 | 89.3 | 69.4 |
| GPT-5.6 Terra xhigh | 97.2 | 87.7 | 72.9 |
| Muse Spark 1.1 xhigh | 98.8 | 89.8 | 69.1 |
| GPT-5.1 Codex max | 98.8 | 89.0 | 68.9 |
| Inkling max | 98.4 | 89.8 | 68.3 |
| GPT-5.5 xhigh | 98.4 | 87.7 | 69.7 |
All results
56 rows| # | Model | Effort | LiveCodeBench (Vals) | Subscription $ / task | API $ / task | Time |
|---|---|---|---|---|---|---|
| 1 | Fable 5.1 | max | 90.5% | $0.041 | $0.26 | 60s |
| 2 | Fable 5 | max | 89.8% | $0.048 | $0.43 | 2.0 min |
| 3 | Gemini 3.8 Flash | high | 89.5% | $0.003 | $0.071 | 24s |
| 4 | Opus 5 | max | 89.0% | $0.002 | $0.13 | 59s |
| 5 | Gemini 3.7 Flash | high | 88.7% | $0.005 | $0.040 | 15s |
| 6 | Grok 4.6 | high | 88.2% | $0.004 | $0.040 | 2.0 min |
| 7 | Gemini 3.6 Flash | high | 88.1% | $0.002 | $0.056 | 32s |
| 8 | GPT-5.2 Codex | xhigh | 88.0% | $0.12 | $0.12 | 2.8 min |
| 9 | Qwen 3.8 Max | max | 87.9% | $0.011 | $0.091 | 5.6 min |
| 10 | Opus 4.8 | max | 87.8% | $0.006 | $0.33 | 2.7 min |
| 11 | Gemini 3.5 Flash | high | 87.6% | $0.086 | $0.086 | 64s |
| 12 | DeepSeek V4 Pro 0813 | max | 87.5% | $0.064 | $0.064 | 4.6 min |
| 13 | Grok 4.5 | high | 87.4% | $0.007 | $0.045 | 9.8 min |
| 14 | GPT-5.3 Codex | xhigh | 87.3% | $0.092 | $0.092 | 2.2 min |
| 15 | DeepSeek V4 Flash 0731 | high | 87.3% | $0.020 | $0.020 | 2.1 min |
| 16 | Kimi K3 | max | 87.2% | $0.009 | $0.10 | 3.3 min |
| 17 | Qwen 3.7 Max | max | 87.1% | $0.026 | $0.087 | 3.2 min |
| 18 | Kimi K2.6 | max | 86.8% | $0.033 | $0.065 | 8.8 min |
| 19 | Qwen 3.6 Plus | max | 86.0% | $0.039 | $0.078 | 7.9 min |
| 20 | Inkling Small | max | 85.9% | $0.017 | $0.021 | 5.0 min |
| 21 | GPT-5.6 Terra | xhigh | 85.9% | $0.001 | $0.031 | 44s |
| 22 | Muse Spark 1.1 | xhigh | 85.9% | $0.046 | $0.046 | 1.9 min |
| 23 | GPT-5.1 Codex | max | 85.5% | $0.14 | $0.14 | 3.9 min |
| 24 | Inkling | max | 85.5% | $0.014 | $0.15 | 4.3 min |
| 25 | GPT-5.5 | xhigh | 85.3% | $0.002 | $0.067 | 2.8 min |
| 26 | Opus 4.7 | max | 85.1% | $0.35 | $0.35 | 7.7 min |
| 27 | GPT-5 Codex | high | 84.7% | $0.097 | $0.097 | 2.2 min |
| 28 | Grok 4.3 | high | 84.5% | $0.020 | $0.020 | 54 min |
| 29 | Qwen 3.8 27B | xhigh | 84.0% | $0.003 | $0.045 | 2.5 min |
| 30 | Qwen 3.5 Flash | max | 83.3% | $0.010 | $0.010 | 3.5 min |