Vals-hosted Terminal-Bench 1.0 mirror (Vals Terminal-Bench 1.0 mirror)

We show this table for reference; we do not rank on it.

Data verified January 12, 202634 confirmed releases in the last 30 daysFollow model changes

A Vals-hosted view of Terminal-Bench 1.0 with easy, medium, and hard task splits.

Vals Terminal-Bench 1.0 mirror score on Vals Terminal-Bench 1.0 mirror — January 12, 2026

We mirror the published vals terminal-bench 1.0 mirror score view for Vals Terminal-Bench 1.0 mirror. GPT-5.2 leads the public snapshot at 63.75%, followed by Claude Sonnet 4.5 Thinking (61.25%) and Gemini 3 Flash Preview (60.00%). We do not use these results to rank models overall.

1Proprietary

GPT-5.2

OpenAI

xhigh reasoning

63.75%

Overall 61.37Context 400K

2Proprietary

Claude Sonnet 4.5 Thinking

Anthropic

61.25%

Context 200K

Gemini 3 Flash Preview

Google

high reasoning

60.00%

47 modelsAgenticCurrentDisplay onlyUpdated January 12, 2026

Vals Terminal-Bench 1.0 mirror score table (47 models)

Score

GPT-5.2OpenAI · Closedxhigh reasoning

63.75%

Claude Sonnet 4.5 ThinkingAnthropic · Closed

61.25%

Gemini 3 Flash PreviewGooglehigh reasoning

60.00%

GPT-5 CodexOpenAIhigh reasoning

58.75%

Claude Opus 4.5 ThinkingAnthropic · Closed

57.50%

GPT-5.1-CodexOpenAI · Closedhigh reasoning

57.50%

Claude Opus 4.5Anthropic · Closed

56.25%

GPT-5.1-Codex-MaxOpenAI · Closedhigh reasoning

53.75%

Gemini 3 Pro PreviewGooglehigh reasoning

51.25%

Claude Haiku 4.5 ThinkingAnthropic · Closed

50.00%

DeepSeek V3p2Fireworks AInone reasoning

50.00%

GLM-4.7Z.AI · Open weight

50.00%

GPT-5OpenAIhigh reasoning

48.75%

GPT-5.1OpenAI · Closedhigh reasoning

47.50%

Claude Sonnet 4 20250514 ThinkingAnthropic

45.00%

Devstral 2512Mistral AI

43.75%

GLM-4.6Z.AI · Open weight

42.50%

GLM-4.5Z.AI · Closed

41.25%

Gemini 2.5 ProGoogle · Closed

41.25%

MiniMax M2.1MiniMax

41.25%

DeepSeek V3p1Fireworks AI

41.25%

Kimi K2 ThinkingMoonshot AI

40.00%

Moonshotai Kimi K2 InstructTogether AI

40.00%

Labs Devstral Small 2512Mistral AI

40.00%

DeepSeek V3p2 ThinkingFireworks AIhigh reasoning

40.00%

Grok 4 0709xAI

38.75%

Kimi K2 Instruct 0905Fireworks AI

37.50%

Qwen3 Max PreviewAlibaba

36.25%

Qwen3 MaxAlibaba · Closed

36.25%

GPT-4.1OpenAI · Closedhigh reasoning

33.75%

Gemini 2.5 Flash Preview 09 2025Google

31.25%

GPT-5 miniOpenAI · Closedhigh reasoning

30.00%

Grok 4.1 Fast (Reasoning)xAI · Closed

28.75%

Magistral Medium 2509Mistral AI

28.75%

Grok 4 Fast (Reasoning)xAI · Closed

27.50%

Gemini 2.5 Flash Preview 09 2025 ThinkingGoogle

26.25%

GPT-OSS 120BOpenAI · Open weight

22.50%

Grok 4.1 Fast Non ReasoningxAI

21.25%

Mistral Large 2512Mistral AI

21.25%

Gemini 2.5 Flash ThinkingGoogle

21.25%

Grok Code Fast 1xAI · Closed

20.00%

Grok 4 Fast Non ReasoningxAI

18.75%

Llama4 Maverick Instruct BasicFireworks AI

17.50%

Magistral Small 2509Mistral AI

15.00%

DeepSeek-R1DeepSeek · Open weight

13.75%

Command A 03 2025Cohere

6.25%

Jamba Large 1.7AI21 Labs

6.25%

How Vals Terminal-Bench 1.0 mirror is shown here

BenchLM mirrors the public Vals AI Vals Terminal-Bench 1.0 mirror leaderboard captured from https://www.vals.ai/benchmarks/terminal-bench and updated by Vals on January 12, 2026. The snapshot preserves overall scores, uncertainty, latency, cost-per-test metadata, and task-level scores where Vals publishes them.

Vals Terminal-Bench 1.0 mirror is display only on BenchLM. Vals proprietary or Vals-hosted aggregate views are useful context, but BenchLM does not use them as weighted ranking inputs or as a replacement for benchmark-native source records.

Snapshot

47 Vals rows4 task viewspublic datasetTasks: Overall, Easy, Medium, HardDisplay only

Vals Terminal-Bench 1.0 mirror on Vals AI Vals methodology Vals home

The published Vals Terminal-Bench 1.0 mirror snapshot places GPT-5.2 first at 63.75%. The third row is 3.75 points behind. The broader top-10 range is 13.75 points, so the table still separates the published systems.

47 models have been evaluated on Vals Terminal-Bench 1.0 mirror. The benchmark falls in the Agentic category. We keep external benchmark mirrors separate from the weighted global scoring system, so these results remain source-specific evidence. Vals Terminal-Bench 1.0 mirror is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About Vals Terminal-Bench 1.0 mirror

Year

2026

Tasks

Terminal tasks split by easy, medium, and hard difficulty

Format

Accuracy score

Difficulty

Terminal-agent execution

We mirror the 47-row Vals table as historical, display-only evidence. It stays separate from current Terminal-Bench 2.x and 3.0 results because the benchmark version and evaluation setup differ.

Terminal-Bench 1.0 Public benchmark source

Freshness and provenance

Version

Vals Terminal-Bench 1.0 mirror 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Questions

What does Vals Terminal-Bench 1.0 mirror measure?

A Vals-hosted view of Terminal-Bench 1.0 with easy, medium, and hard task splits.

Which model leads the published Vals Terminal-Bench 1.0 mirror snapshot?

GPT-5.2 currently leads the published Vals Terminal-Bench 1.0 mirror snapshot with 63.75% vals terminal-bench 1.0 mirror score. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on Vals Terminal-Bench 1.0 mirror?

The January 12, 2026 snapshot contains 47 AI models.

Last updated: January 12, 2026 · mirrored from the public benchmark leaderboard

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.