Skip to content
Hbench

Results

Results

Every published result, one row per harness version on each board. Filter it, or download what you see as CSV.

56 results

Download CSV

Select two rows of one board to compare them task by task.

Results, one row per configuration on each board
HarnessBoardPass rateHigher is better, 95% intervalRunsCost per taskhbenchLast finished
Terminal-Bench 2.0 on GPT-5.5 (5 attempts, open network)83.1%76.6–89.21—converted28 Apr 2026
clnkr0.3.11
Terminal-Bench 2.0 on GPT-5.5 (5 attempts, open network)66.1%57.5–74.21—converted7 May 2026
Terminal-Bench 2.0 on MiniMax-M2.7 (5 attempts, open network)34.4%27.4–40.91—converted22 Mar 2026
Terminal-Bench 2.0 on GPT-5.3-Codex (5 attempts, open network)77.3%69.4–84.51—converted22 Feb 2026
Terminal-Bench 2.0 on GPT-5.3-Codex (5 attempts, open network)74.6%67.0–81.61—converted12 Feb 2026
Terminal-Bench 2.0 on GPT-5.3-Codex (5 attempts, open network)71.5%63.1–79.11$2.44converted17 Mar 2026
Terminal-Bench 2.0 on GPT-5.3-Codex (5 attempts, open network)58.7%50.6–65.81—converted17 Mar 2026
Terminal-Bench 2.0 on Claude Opus 4.6 (5 attempts, open network)82.0%74.2–88.51—converted28 Mar 2026
Terminal-Bench 2.0 on Claude Opus 4.6 (5 attempts, open network)76.4%69.2–83.41$2.06converted26 Mar 2026
maya-v21.5-official
Terminal-Bench 2.0 on Claude Opus 4.6 (5 attempts, open network)71.9%63.4–79.61—converted9 Mar 2026
Judy0.7.0
Terminal-Bench 2.0 on Claude Opus 4.6 (5 attempts, open network)71.9%63.8–79.31—converted13 Feb 2026
Terminal-Bench 2.0 on Claude Opus 4.6 (5 attempts, open network)69.9%61.8–77.31—converted30 Jan 2026
wozcode2.1.86
Terminal-Bench 2.0 on Claude Opus 4.6 (5 attempts, open network)68.1%60.0–76.41$0.75converted29 Mar 2026
Terminal-Bench 2.0 on Claude Opus 4.6 (5 attempts, open network)62.9%54.4–70.81$0.66converted6 Feb 2026
Terminal-Bench 2.0 on Claude Opus 4.6 (5 attempts, open network)60.4%51.5–68.81—converted3 Apr 2026
final1.5-official
Terminal-Bench 2.0 on Claude Sonnet 4.5 (5 attempts, open network)42.7%32.6–52.81—converted9 Dec 2025
Terminal-Bench 2.0 on claude-opus-4-558.4%48.3–68.51—converted16 Jan 2026
Terminal-Bench 2.0 on claude-opus-4-551.7%41.6–61.81—converted12 Jan 2026
Terminal-Bench 2.0 on gpt-5.260.7%50.6–70.81—converted16 Jan 2026
Terminal-Bench 2.0 on claude-sonnet-4-6 (5 attempts, open network)52.1%43.6–60.71—converted17 Apr 2026
aone-agent1.18.72-pre
Terminal-Bench 2.0 on qwen3-coder-480b-a35b-instruct-fp8 (5 attempts, open network)27.2%19.8–35.11—converted26 Dec 2025
Terminal-Bench 2.0 on minimax-m2.7-highspeed (5 attempts, open network)42.5%33.9–50.81—converted2 May 2026
Terminal-Bench 2.0 on glm-5 (5 attempts, open network)47.4%39.1–55.51—converted5 Apr 2026
Terminal-Bench 2.0 on qwen3.5-9b (5 attempts, open network)9.2%5.8–13.31—converted27 Apr 2026
Terminal-Bench 2.0 on qwen3.6-35b-a3b (5 attempts, open network)23.8%17.1–31.21—converted26 Apr 2026