BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases7

Terminal-Bench 2.0 (Vals) leaderboard

As of 6 Nov 2026, the top score on Terminal-Bench 2.0 (Vals) is 73.2% by GPT-5.5 (OpenAI), ahead of Claude Opus 4.8 at 70.0%, across 64 model configurations with a published result (Vals AI, result dated 4 Jun 2026). Terminal-Bench 2.0: agent tasks in a shell, checked by tests, run by Vals AI. Terminal-Bench 2.0 (Vals) is shown for reference only and does not count toward the BenchLeader Index.

Key facts (as of 6 Nov 2026)

  • GPT-5.5 holds the top Terminal-Bench 2.0 (Vals) score at 73.2% (Vals AI, 4 Jun 2026).
  • Claude Opus 4.8 is second on Terminal-Bench 2.0 (Vals) at 70.0% (Vals AI, 4 Jun 2026).
  • Terminal-Bench 2.0 (Vals): Terminal-Bench 2.0 in Vals AI's fixed harness. Run by Vals AI (Vals AI).
  • 64 model configurations have a published Terminal-Bench 2.0 (Vals) result on BenchLeader (BenchLeader, 6 Nov 2026).
  • Terminal-Bench 2.0 (Vals) does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).

Terminal-Bench 2.0 in Vals AI's fixed harness. Run by Vals AI.

Published by
Vals AI
Category
Agents & tools
Index weight
Reference only
Models
64
Data as of
6 Nov 2026

Vals AI (vals.ai).

What the test looks like

Terminal-Bench 2.0: agent tasks in a shell, checked by tests, run by Vals AI.

How it is scored

Percent of tasks passing, run by Vals AI.

What to keep in mind

Superseded by 2.1, kept for comparison.

Top 15
  1. 1GPT-5.5 (high)73.2%
  2. 2Claude Opus 4.870.0%
  3. 3Claude Opus 4.768.5%
  4. 4Gemini 3.5 Flash (high)67.4%
  5. 5Gemini 3.1 Pro (high)67.4%
  6. 6GPT-5.3-Codex (high)64.0%
  7. 7Muse Spark59.5%
  8. 8Claude Sonnet 4.659.5%
  9. 9Qwen3.7 (max)59.2%
  10. 10Claude Opus 4.6 (thinking)58.4%
  11. 11GPT-5.4 (high)58.4%
  12. 12Claude Opus 4.558.4%
  13. 13Kimi K2.657.3%
  14. 14DeepSeek V4 Pro56.2%
  15. 15Gemini 3 Pro (high)55.1%
64 of 64
#
1GPT-5.5high73.2%63.7$0.7612026-06-04
2Claude Opus 4.870.0%60.0$0.8842026-06-04
3Claude Opus 4.768.5%63.1$0.9762026-06-04
4Gemini 3.5 Flashhigh67.4%62.5$1.192026-06-04
5Gemini 3.1 Prohigh67.4%60.1$0.5162026-06-04
6GPT-5.3-Codexhigh64.0%–$0.3452026-06-04
7Muse Spark59.5%64.1$0.0702026-06-04
8Claude Sonnet 4.659.5%58.2$0.5192026-06-04
9Qwen3.7max59.2%60.3$0.7192026-06-04
10Claude Opus 4.6thinking58.4%62.4$1.142026-06-04
11GPT-5.4high58.4%58.8$0.5122026-06-04
12Claude Opus 4.558.4%57.9$1.182026-06-04
13Kimi K2.657.3%55.2$0.1932026-06-04
14DeepSeek V4 Pro56.2%53.2$0.1802026-06-04
15Gemini 3 Prohigh55.1%60.1$0.4562026-06-04
16Claude Opus 4.5thinking53.9%59.6$1.122026-06-04
17GLM 5.153.9%55.7$0.1582026-06-04
18Qwen3.6 Maxmax51.7%62.4$0.3202026-06-04
19Gemini 3 Flashhigh51.7%58.3$0.1692026-06-04
20GPT-5.2high51.7%56.0$0.4702026-06-04
21GLM-5thinking49.4%60.2$0.2472026-06-04
22MiniMax M2.747.2%56.5$0.0742026-06-04
23MiniMax M346.1%56.4$0.2162026-06-04
24Qwen3.6 Plus44.9%59.1$0.1462026-06-04
25GPT-5.1high44.9%57.7$0.4082026-06-04
26GPT-5.4 Minihigh44.9%52.1$0.3052026-06-04
27Qwen3.6 27B44.9%48.0$0.2982026-06-04
28Grok 4.343.5%54.5$0.3732026-06-04
29Qwen3.5 Plusthinking41.6%57.3$0.3222026-06-04
30MiniMax M2.541.6%53.6$0.0852026-06-04
31Claude Sonnet 4.5thinking41.6%52.2$0.7512026-06-04
32Kimi K2.5thinking40.5%58.7$0.1402026-06-04
33Grok 4.20thinking40.5%58.4$0.1542026-06-04
34GPT-5.4 Nanohigh39.9%48.3$0.0752026-06-04
35Gemma 4 31Bhigh39.3%––2026-06-04
36GLM 4.738.2%51.7$0.1842026-06-04
37Claude Haiku 4.5thinking38.2%45.4$0.3282026-06-04
38GPT-5high37.1%57.9$0.4772026-06-04
39MiniMax-M2.137.1%55.0$0.0702026-06-04
40Kimi K2 Thinkingthinking37.1%53.2$0.1162026-06-04
41DeepSeek V3.2high36.0%–$0.1662026-06-04
42Laguna M.131.5%37.3–2026-06-04
43Gemini 2.5 Pro30.3%54.5$0.4722026-06-04
44Mistral Medium 3.5high30.3%–$4.712026-06-04
45Grok 4 Fastthinking29.2%51.6$0.0342026-06-04
46Grok 428.1%58.3$0.5532026-06-04
47GLM-4.628.1%49.8$0.1442026-06-04
48Laguna XS.228.1%36.3–2026-06-04
49GPT-5 minihigh27.0%54.6$0.1172026-06-04
50Kimi K225.8%51.0$1.112026-06-04
51Grok 4.1 Fastthinking24.7%55.6$0.0562026-06-04
52Qwen3 Maxmax24.7%54.3$0.4232026-06-04
53Qwen3.5 Flash24.7%51.8$0.1102026-06-04
54Gemini 3.1 Flash Litehigh24.7%48.5$0.0652026-06-04
55Gemini 2.5 Flash 09 2025thinking21.4%52.8$0.1442026-06-04
56gpt-oss-120b19.1%45.5$0.1322026-06-04
57Trinity Largethinking18.0%47.1$0.3262026-06-04
58Grok 4.1 Fastno reasoning18.0%43.5$0.0332026-06-04
59Mistral Small 4high16.9%–$0.1062026-06-04
60GPT-4.1high14.6%46.6$1.082026-06-04

More agents & tools benchmarks · All LLM benchmarks

Related benchmarks

Top head-to-heads on Terminal-Bench 2.0 (Vals)

Cite as: BenchLeader, “Terminal-Bench 2.0 (Vals) leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/vals_terminal_bench_2.

Terminal-Bench 2.0 (Vals): questions

What does Terminal-Bench 2.0 (Vals) measure?
Terminal-Bench 2.0: agent tasks in a shell, checked by tests, run by Vals AI. Scores are reported in percent of tasks solved; higher is better.
What is the top score on Terminal-Bench 2.0 (Vals)?
As of 6 Nov 2026, the highest score on Terminal-Bench 2.0 (Vals) is 73.2% by GPT-5.5 (result dated 4 Jun 2026), ahead of Claude Opus 4.8 at 70.0%. Scores come from Vals AI and are re-read daily.
Which AI model leads Terminal-Bench 2.0 (Vals)?
GPT-5.5 leads Terminal-Bench 2.0 (Vals) with 73.2% as of 6 Nov 2026 (result dated 4 Jun 2026), ahead of Claude Opus 4.8 at 70.0%.
How many models have Terminal-Bench 2.0 (Vals) results?
64 model configurations have a Terminal-Bench 2.0 (Vals) result on BenchLeader, all taken from Vals AI.
Who runs Terminal-Bench 2.0 (Vals) and how often is it updated?
Terminal-Bench 2.0 (Vals) is published by Vals AI. BenchLeader re-reads the published results once a day and records the date each result was published.
Does Terminal-Bench 2.0 (Vals) count toward the BenchLeader Index?
No. Terminal-Bench 2.0 (Vals) is shown for reference but left out of the composite index.