Inference
EN
Artificial Analysis

Terminal-Bench Hard

ModelsModels with a result.23
Top resultBest result on this benchmark.65.9%GPT-5.6 Sol
Top-3 spreadPoints from first to third.5.3 pts
YearYear of release.2026

Result and price

20 of 23 results

Results

23 results
#ModelResultThe score from the source.
01GPT-5.6 Sol65.9%
02Claude Fable 562.9%
03GPT-5.560.6%
04Claude Opus 4.858.3%
05GPT-5.457.6%
06Claude Opus 4.754.5%
07Gemini 3.1 Pro53.8%
08GPT-5.4 mini52.3%
09GLM-5.250.8%
09Qwen3.7 Max50.8%
11Claude Opus 4.648.5%
12GPT-5.247.0%
12Qwen3.7 Plus47.0%
12Claude Opus 4.5 Thinking47.0%
15Muse Spark45.5%
15GPT-5.145.5%
17Kimi K2.7 Code44.7%
18Kimi K2.643.9%
18Qwen 3.6 Max (preview)43.9%
18Qwen3.6 Plus43.9%
21MiMo-V2.5-Pro43.2%
21GLM-543.2%
23Solar Open 228.3%