SShogo Bench

Terminal-Bench 2.0

Hard tasks done entirely in a terminal: compiling, debugging, data wrangling, sysadmin. Best reported agent per model.

Agent Index weight
8%
in the Agentic coding pillar
Models tested
44
from the benchmark’s leaderboard and reports
Normalised for the index
0 → 0, 1 → 100
Official site

Results

#ModelScoreNormalisedHarnessCost/taskDate
1GPT-5.5 Supported
OpenAI
84.7%NexAU-AHE—Apr 23, 2026
2GPT-5.4 Supported
OpenAI
81.8%ForgeCode—Mar 12, 2026
3Claude Opus 4.7 Supported
Anthropic
80.2%WOZCODE—Apr 16, 2026
4Gemini 3.1 Pro Preview Supported
Google
80.2%TongAgents—Feb 19, 2026
5Claude Opus 4.6 Supported
Anthropic
79.8%ForgeCode—Mar 12, 2026
6GPT-5.3 Codex Supported
OpenAI
78.4%SageAgent—Feb 5, 2026
7Gemini 3 Pro Preview Supported
Google
69.4%Ante—Nov 18, 2025
8GPT-5.2 Codex
OpenAI
66.5%Deep Agents—Jan 14, 2026
9GPT-5.2 Supported
OpenAI
64.9%Droid—Dec 11, 2025
10Gemini 3 Flash Preview Supported
Google
64.3%Junie CLI—Dec 17, 2025
11Claude Opus 4.5 (latest) Supported
Anthropic
63.1%Droid—Nov 24, 2025
12GPT-5.1 Codex mini
OpenAI
61.6%hookele—Nov 12, 2025
13GPT-5.1 Codex Max Estimated
OpenAI
60.4%Codex CLI—Nov 19, 2025
14GPT-5.1 Codex
OpenAI
57.8%Codex CLI—Nov 16, 2025
15Grok 4.20 (Non-Reasoning) Estimated
xAI
57.3%Grok CLI—Mar 10, 2026
16Claude Sonnet 4.6 Supported
Anthropic
53.4%Simplai Agent—Feb 17, 2026
17GLM-5 Supported
Z.ai (Zhipu)
52.4%Terminus 2—Feb 11, 2026
18GPT-5 Supported
OpenAI
49.6%Codex CLI—Aug 7, 2025
19GPT-5.1 Supported
OpenAI
47.6%Terminus 2—Nov 16, 2025
20Claude Sonnet 4.5 (latest) Supported
Anthropic
46.5%CAMEL-AI—Sep 29, 2025
21MiniMax-M2.7
MiniMax
45.1%IndusAGI Coding Agent—Mar 18, 2026
22GPT-5-Codex
OpenAI
44.3%Codex CLI—Sep 15, 2025
23Kimi K2.5 Supported
Moonshot AI
43.2%Terminus 2—Jan 27, 2026
24MiniMax-M2.5 Estimated
MiniMax
42.7%cchuter—Feb 12, 2026
25DeepSeek V3.2 Supported
DeepSeek
39.6%Terminus 2—Dec 1, 2025
26Claude Opus 4.1 (latest) Supported
Anthropic
38.0% Terminus 2 —Oct 31, 2025
27MiniMax-M2.1
MiniMax
36.6%Crux—Dec 23, 2025
28Kimi K2 Thinking Estimated
Moonshot AI
35.7%Terminus 2—Nov 6, 2025
29Claude Haiku 4.5 Estimated
Anthropic
35.5%Goose—Oct 15, 2025
30GPT-5 Mini Estimated
OpenAI
34.8%spoox-o-m—Aug 7, 2025
31GLM-4.7 Estimated
Z.ai (Zhipu)
33.4%Terminus 2—Dec 22, 2025
32Gemini 2.5 Pro Supported
Google
32.6%Terminus 2—Jun 17, 2025
33MiniMax-M2 Estimated
MiniMax
30.0%Terminus 2—Oct 27, 2025
34Kimi K2 0905 Estimated
Moonshot AI
27.8% Terminus 2 —Nov 1, 2025
35Grok 4 Supported
xAI
27.2% OpenHands —Nov 2, 2025
36Qwen3 Coder 480B A35B Estimated
Alibaba (Qwen)
27.2%Dakou Agent—Jul 22, 2025
37Grok Code Fast 1
xAI
25.8%Mini-SWE-Agent—Aug 28, 2025
38Qwen3.6 35B-A3B
Alibaba (Qwen)
24.6%little-coder—Apr 16, 2026
39GLM-4.6
Z.ai (Zhipu)
24.5%Terminus 2—Sep 30, 2025
40GPT-5 Nano
OpenAI
21.8%spoox-o-m—Aug 7, 2025
41GPT OSS 120B Supported
OpenAI
18.7% Terminus 2 —Nov 1, 2025
42Gemini 2.5 Flash Estimated
Google
17.1% Mini-SWE-Agent —Nov 3, 2025
43Qwen3.5 9B
Alibaba (Qwen)
9.2%little-coder—Mar 2, 2026
44GPT OSS 20B
OpenAI
3.4% Mini-SWE-Agent —Nov 3, 2025

For models without a Terminal-Bench 2.0 result, the index estimates one from Epoch’s Capabilities Index using a logistic fit on 31 models (residual SD 0.38 on the logit scale). Estimates are labelled and widen the model’s range.