LiveFMBench - Pre2025: leaderboard
The historical split of LiveFMBench (2026): 270 C programs from earlier ACSL benchmarks and GitHub, for which a model must write ACSL specifications (preconditions, postconditions, loop invariants) that let Frama-C prove the embedded assertions. The score is pass@1 (%) over 32 samples under direct prompting, counting a sample only when the proof succeeds and the model left the original code and assertions unchanged; higher is better.
Metric: Pass@1 (%) after the faithfulness check on LiveFMBench-Pre2025 (270 C programs from AutoSpec, FMBench and GitHub), direct prompting (instructions plus the C program), 32 samples at temperature 0.7 and top-p 0.95, pass@1 estimated from the 32 samples with the unbiased estimator; a sample counts only when Frama-C/WP (Alt-Ergo and Z3) proves the program and its code and assertions are unchanged (faithfulness check); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 11 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Qwen 3 235B A22B 2507 (Thinking) | 34.92 | #471 | |
| 2 | DeepSeek R1 0528 | 34.69 | #432 | |
| 3 | DeepSeek V3.1 Terminus | Thinking | 30.52 | #442 |
| 4 | DeepSeek V3.2 | Thinking | 30.31 | #425 |
| 5 | Kimi K2 0905 | 27.49 | #578 | |
| 6 | Qwen 3 32B | 27.44 | #983 | |
| 7 | GLM-4.6 | 27.08 | #496 | |
| 8 | Qwen 3 Coder 480B A35B Instruct | 27.04 | #655 | |
| 9 | DeepSeek V3 (0324) | 19.47 | #730 | |
| 10 | Qwen 3 235B A22B 2507 Instruct | 11.54 | #625 | |
| 11 | Llama 3.3 70B Instruct | 4.27 | #1394 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=livefmbench-pre2025 · How It Works · Data refreshed daily, snapshot 2026-11-22.