llmboard.aiAI model intelligence
Home

Model Rankings

OverallOpen ModelsAgentCodingReasoningMathKnowledgeInstruction FollowingTextVision
Image GenerationImage Editing
Video GenerationImage to VideoVideo Editing
Text to SpeechSpeech to Text
Embeddings

Efficiency

Chat Token PricingImage PricingVideo PricingAudio Pricing
Chat Speed & LatencyProvider Reliability

Benchmarks

GPQAMMLU-ProAIME 2025SWE-Bench VerifiedMMLUHumanity's Last ExamLiveCodeBenchMATHHumanEvalMMMU-Pro
All Benchmarks

Tools

Model Directory

Scoring & Data

Scoring & Data
1367 models808 benchmarks

Leaderboard Center

Overall RankingCodingCore BenchmarksPrice & ValueRuntime Performance

Modalities

All ModelsImage GenerationImage EditingVideo GenerationImage-to-VideoVideo EditingText-to-SpeechSpeech-to-TextEmbeddings

Data & Methods

Scoring MethodAll Benchmarks
llmboard.aiCopyright 2026 llmboard.ai

reasoning benchmark

Terminal-Bench 2.1 Leaderboard

Terminal-Bench 2.1 is an updated benchmark that tests AI agents’ ability to operate a computer via the terminal.

Updated Oct 23, 2026

Models48
Model coverage48
MetricScore
EvidenceB

On this page

  • Ranking
  • Highlights
  • Distribution
  • Top models
  • About
  • FAQ

Terminal-Bench 2.1 Ranking

Higher score ranks better on this benchmark.

30 of 48 rows
Columns

Show columns

Sort by
Rank
Model
Score
Percentile
Participants
Evidence
Evaluated
Rank01ModelDEDeepSeek-V4.1-FlashDeepSeekScore90.60%Percentile100.00%Participants48EvidenceCEvaluatedOct 22, 2026
Rank02ModelXIMiMo-V2.6-ProXiaomiScore89.90%Percentile97.87%Participants48EvidenceCEvaluatedOct 22, 2026
Rank03ModelGOGemini 3.8 FlashGoogleScore89.40%Percentile95.74%Participants48EvidenceCEvaluatedOct 22, 2026
Rank04ModelOPGPT-5.6 SolOpenAIScore88.80%Percentile93.62%Participants48EvidenceCEvaluatedOct 22, 2026
Rank05ModelMEMuse Spark 1.3MetaScore88.80%Percentile91.49%Participants48EvidenceCEvaluatedOct 22, 2026
Rank06ModelMAKimi K3Moonshot AIScore88.30%Percentile89.36%Participants48EvidenceCEvaluatedOct 22, 2026
Rank07ModelZAGLM-5.3Zhipu AIScore88.20%Percentile87.23%Participants48EvidenceCEvaluatedOct 22, 2026
Rank08ModelDEDeepSeek-V4-Pro-0813DeepSeekScore87.90%Percentile85.11%Participants48EvidenceCEvaluatedOct 22, 2026
Rank09ModelXIMiMo-V2.6-FlashXiaomiScore87.60%Percentile82.98%Participants48EvidenceCEvaluatedOct 22, 2026
Rank10ModelOPGPT-5.6 TerraOpenAIScore87.40%Percentile80.85%Participants48EvidenceCEvaluatedOct 22, 2026
Rank11ModelACQwen3.8 MaxAlibaba Cloud / Qwen TeamScore86.60%Percentile78.72%Participants48EvidenceCEvaluatedOct 22, 2026
Rank12ModelGOGemini 3.7 FlashGoogleScore85.80%Percentile76.60%Participants48EvidenceCEvaluatedOct 22, 2026
Rank13ModelTEHy4 previewTencentScore85.40%Percentile74.47%Participants48EvidenceCEvaluatedOct 22, 2026
Rank14ModelOPGPT-5.6 LunaOpenAIScore84.70%Percentile72.34%Participants48EvidenceCEvaluatedOct 22, 2026
Rank15ModelANClaude Fable 5AnthropicScore84.30%Percentile70.21%Participants48EvidenceCEvaluatedOct 22, 2026
Rank16ModelZAGLM-5.3-FlashZhipu AIScore84.30%Percentile68.09%Participants48EvidenceCEvaluatedOct 22, 2026
Rank17ModelDEDeepSeek-V4-Flash-Vision-ExpDeepSeekScore83.90%Percentile65.96%Participants48EvidenceCEvaluatedOct 22, 2026
Rank18ModelXAGrok 4.5xAIScore83.30%Percentile63.83%Participants48EvidenceCEvaluatedOct 22, 2026
Rank19ModelMEMuse Spark 1.2MetaScore82.90%Percentile61.70%Participants48EvidenceCEvaluatedOct 22, 2026
Rank20ModelDEDeepSeek-V4-Flash-0731DeepSeekScore82.70%Percentile59.57%Participants48EvidenceCEvaluatedOct 22, 2026
Rank21ModelZAGLM-5.2Zhipu AIScore82.70%Percentile57.45%Participants48EvidenceCEvaluatedOct 22, 2026
Rank22ModelFAEmber-1Fireworks AIScore82.00%Percentile55.32%Participants48EvidenceCEvaluatedOct 22, 2026
Rank23ModelRABeam 1.0Reflection AIScore80.10%Percentile53.19%Participants48EvidenceCEvaluatedOct 22, 2026
Rank24ModelMEMuse Spark 1.1MetaScore80.00%Percentile51.06%Participants48EvidenceCEvaluatedOct 22, 2026
Rank25ModelSAAtria Dawn PreviewShanghai AI LaboratoryScore78.30%Percentile48.94%Participants48EvidenceCEvaluatedOct 12, 2026
Rank26ModelGOGemini 3.6 FlashGoogleScore78.00%Percentile46.81%Participants48EvidenceCEvaluatedOct 22, 2026
Rank27ModelMEMuse Spark OpenMetaScore76.40%Percentile44.68%Participants48EvidenceCEvaluatedOct 22, 2026
Rank28ModelACQwen3.8-27BAlibaba Cloud / Qwen TeamScore73.00%Percentile42.55%Participants48EvidenceCEvaluatedOct 22, 2026
Rank29ModelTEHy3TencentScore71.70%Percentile40.43%Participants48EvidenceCEvaluatedOct 22, 2026
Rank30ModelBYSeed 2.1 ProByteDanceScore71.00%Percentile38.30%Participants48EvidenceCEvaluatedOct 22, 2026

Terminal-Bench 2.1 Highlights

The leading models and scores on this benchmark.

Terminal-Bench 2.1 Score Distribution

A closer view of the leading scores on this benchmark.

Terminal-Bench 2.1

The Top AI Models for Terminal-Bench 2.1

The first five results on this benchmark, with official price and output speed added where the model identity can be matched.

What is Terminal-Bench 2.1?

What Terminal-Bench 2.1 measures and how its scores work.

Terminal-Bench 2.1 is a benchmark for evaluating AI agents on real-world, end-to-end computer tasks performed through the terminal.

It measures how well models autonomously handle tasks including compiling code, training models, setting up servers, system administration, data science workflows, and security tasks, reported as a Score ratio.

Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.

Family
Terminal-Bench 2.1
Modality
text
Primary category
reasoning
Score direction
higher
LLMBoard eligible
No
Evaluation key
overall

LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.

FAQ

Common questions about Terminal-Bench 2.1.

Which model scores highest on Terminal-Bench 2.1?

DeepSeek-V4.1-Flash is currently ranked first with 90.60%.

What are the top three models on Terminal-Bench 2.1?

The current leaders are DeepSeek-V4.1-Flash (90.60%), MiMo-V2.6-Pro (89.90%), and Gemini 3.8 Flash (89.40%).

Which Terminal-Bench 2.1 model has the lowest official input price?

GLM-5.3-Flash has the lowest matched official input price at $0.12 input / $0.42 output per 1M tokens.

Which models are fastest among Terminal-Bench 2.1 results?

The fastest matched records are Gemini 3.7 Flash (409.96 tok/s via Google), GPT-5.6 Luna (215.52 tok/s via OpenAI), and Grok 4.5 (80.00 tok/s via xAI).

Does the highest score result prove overall model quality?

No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.

What does Terminal-Bench 2.1 measure?

It measures how well models autonomously handle tasks including compiling code, training models, setting up servers, system administration, data science workflows, and security tasks, reported as a Score ratio.

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

48 model results are currently shown.

Does this benchmark affect the overall score?

No. This benchmark is shown for reference but does not contribute to the overall score.

Rank #1DeepSeek-V4.1-Flash90.60%
Rank #2MiMo-V2.6-Pro89.90%
Rank #3Gemini 3.8 Flash89.40%
Rank #4GPT-5.6 Sol88.80%

Ranking basisThis terminal-bench 2.1 AI model leaderboard uses descending score in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.

  1. 01
    DE
    DeepSeek-V4.1-FlashDeepSeek
    Score
    90.60%
    Price
    $0.15 input / $0.60 output per 1M tokens

    Strengths

    • Ranks #1 of 48 compared models
    • 100th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Terminal-Bench 2.1, not total model capability
  2. 02
    XI
    MiMo-V2.6-ProXiaomi
    Score
    89.90%
    Price
    $0.44 input / $0.87 output per 1M tokens
    Speed
    Up to 13.54 tok/s via Xiaomi

    Strengths

    • Ranks #2 of 48 compared models
    • 98th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Terminal-Bench 2.1, not total model capability
  3. 03
    GO
    Google
    Score
    89.40%
    Price
    $0.75 input / $3.8 output per 1M tokens
    Speed
    Up to 55.66 tok/s via Google

    Strengths

    • Ranks #3 of 48 compared models
    • 96th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Terminal-Bench 2.1, not total model capability
  4. 04
    OP
    OpenAI
    Score
    88.80%
    Price
    $4.0 input / $20 output per 1M tokens
    Speed
    Up to 25.46 tok/s via OpenAI

    Strengths

    • Ranks #4 of 48 compared models
    • 94th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Terminal-Bench 2.1, not total model capability
  5. 05
    ME
    Meta
    Score
    88.80%
    Price
    $1.3 input / $4.3 output per 1M tokens

    Strengths

    • Ranks #5 of 48 compared models
    • 91th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Terminal-Bench 2.1, not total model capability

Selection summary

Best AI Models for Terminal-Bench 2.1

DeepSeek-V4.1-Flash currently leads Terminal-Bench 2.1 with 90.60%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price, and runtime.

Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.

Gemini 3.8 Flash
GPT-5.6 Sol
Muse Spark 1.3
Benchmark rank #1DeepSeek-V4.1-Flash90.60% · $0.15 input / $0.60 output per 1M tokens
Benchmark rank #2MiMo-V2.6-Pro89.90% · $0.44 input / $0.87 output per 1M tokens
Benchmark rank #3Gemini 3.8 Flash89.40% · $0.75 input / $3.8 output per 1M tokens