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
1438 models809 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

Unspecified benchmark

LM Arena Agent Steerability Leaderboard

LM Arena Agent Steerability is presented as a current leaderboard.

Updated Nov 13, 2026

Models43
Model coverage43
MetricAgent Score
EvidenceA

On this page

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

LM Arena Agent Steerability Ranking

Higher agent score ranks better on this benchmark.

30 of 43 rows
Columns

Show columns

Sort by
Rank
Model
Agent Score
Percentile
Participants
Observations
Evidence
Evaluated
Rank01ModelGOGemini 4 CarbonGoogleAgent Score19.06% (18.45% - 19.68%)Percentile100.00%Participants43Observations278.2KEvidenceAEvaluatedNov 11, 2026
Rank02ModelOPGPT-6 AstraOpenAIAgent Score14.82% (13.03% - 16.61%)Percentile97.62%Participants43Observations49.4KEvidenceAEvaluatedNov 11, 2026
Rank03ModelANClaude Fable 5.1AnthropicAgent Score14.80% (12.96% - 16.64%)Percentile95.24%Participants43Observations75.6KEvidenceAEvaluatedNov 11, 2026
Rank04ModelGOGemini 4 ArgonGoogleAgent Score13.16% (12.18% - 14.13%)Percentile92.86%Participants43Observations266KEvidenceAEvaluatedNov 11, 2026
Rank05ModelTMInkling 2 SmallThinking Machines LabAgent Score9.28% (7.50% - 11.07%)Percentile90.48%Participants43Observations53.5KEvidenceAEvaluatedNov 11, 2026
Rank06ModelTEHy4 previewTencentAgent Score8.78% (7.25% - 10.30%)Percentile88.10%Participants43Observations83.9KEvidenceAEvaluatedNov 11, 2026
Rank07ModelANClaude Fable 5AnthropicAgent Score7.88% (6.50% - 9.27%)Percentile85.71%Participants43Observations178.3KEvidenceAEvaluatedNov 11, 2026
Rank08ModelOPGPT-5.6 SolOpenAIAgent Score5.54% (4.11% - 6.97%)Percentile83.33%Participants43Observations173.3KEvidenceAEvaluatedNov 11, 2026
Rank09ModelANClaude Opus 5AnthropicAgent Score4.38% (2.74% - 6.03%)Percentile80.95%Participants43Observations109.7KEvidenceAEvaluatedNov 11, 2026
Rank10ModelXIMiMo-V2.6-FlashXiaomiAgent Score2.85% (1.55% - 4.15%)Percentile78.57%Participants43Observations140KEvidenceAEvaluatedNov 11, 2026
Rank11ModelANClaude Sonnet 5AnthropicAgent Score2.76% (1.05% - 4.47%)Percentile76.19%Participants43Observations114.7KEvidenceAEvaluatedNov 11, 2026
Rank12ModelMEMuse Spark 1.3MetaAgent Score2.63% (1.49% - 3.77%)Percentile73.81%Participants43Observations477.4KEvidenceAEvaluatedNov 11, 2026
Rank13ModelANClaude Haiku 5.5AnthropicAgent Score2.36% (1.04% - 3.69%)Percentile71.43%Participants43Observations298.1KEvidenceAEvaluatedNov 11, 2026
Rank14ModelXIMiMo-V2.6-ProXiaomiAgent Score1.50% (0.31% - 2.69%)Percentile69.05%Participants43Observations228.7KEvidenceAEvaluatedNov 11, 2026
Rank15ModelOPGPT-6 LunaOpenAIAgent Score-0.11% (-0.75% - 0.53%)Percentile66.67%Participants43Observations620.9KEvidenceAEvaluatedNov 11, 2026
Rank16ModelOPGPT-5.6 TerraOpenAIAgent Score-0.52% (-1.64% - 0.60%)Percentile64.29%Participants43Observations234KEvidenceAEvaluatedNov 11, 2026
Rank17ModelACQwen3.8 MaxAlibaba Cloud / Qwen TeamAgent Score-0.78% (-1.76% - 0.21%)Percentile61.90%Participants43Observations390.1KEvidenceAEvaluatedNov 11, 2026
Rank18ModelMAKimi K3Moonshot AIAgent Score-0.79% (-1.65% - 0.08%)Percentile59.52%Participants43Observations562.9KEvidenceAEvaluatedNov 11, 2026
Rank19ModelZAGLM-5.2Zhipu AIAgent Score-1.03% (-2.24% - 0.18%)Percentile57.14%Participants43Observations302.5KEvidenceAEvaluatedNov 11, 2026
Rank20ModelZAGLM-5.3Zhipu AIAgent Score-1.18% (-2.32% - -0.04%)Percentile54.76%Participants43Observations501.4KEvidenceAEvaluatedNov 11, 2026
Rank21ModelGOGemini 3.8 FlashGoogleAgent Score-1.26% (-2.46% - -0.06%)Percentile52.38%Participants43Observations274.9KEvidenceAEvaluatedNov 11, 2026
Rank22ModelXAGrok 4.6xAIAgent Score-1.76% (-3.61% - 0.08%)Percentile50.00%Participants43Observations123.8KEvidenceAEvaluatedNov 11, 2026
Rank23ModelXAGrok 4.7xAIAgent Score-2.27% (-4.28% - -0.25%)Percentile47.62%Participants43Observations100.5KEvidenceAEvaluatedNov 11, 2026
Rank24ModelACQwen3.8-Flash-NextAlibaba Cloud / Qwen TeamAgent Score-2.56% (-3.66% - -1.47%)Percentile45.24%Participants43Observations384.1KEvidenceAEvaluatedNov 11, 2026
Rank25ModelMIMiniMax M3.1 FlashMiniMaxAgent Score-2.57% (-4.66% - -0.47%)Percentile42.86%Participants43Observations51.1KEvidenceAEvaluatedNov 11, 2026
Rank26ModelTEHy3TencentAgent Score-3.09% (-5.07% - -1.10%)Percentile40.48%Participants43Observations53.6KEvidenceAEvaluatedNov 11, 2026
Rank27ModelZAGLM-5.3-FlashZhipu AIAgent Score-5.01% (-6.12% - -3.89%)Percentile38.10%Participants43Observations698.6KEvidenceAEvaluatedNov 11, 2026
Rank28ModelACQwen3.8-27BAlibaba Cloud / Qwen TeamAgent Score-6.00% (-7.42% - -4.58%)Percentile35.71%Participants43Observations266.2KEvidenceAEvaluatedNov 11, 2026
Rank29ModelMAMistral Large 4Mistral AIAgent Score-6.39% (-9.60% - -3.18%)Percentile33.33%Participants43Observations27.1KEvidenceAEvaluatedNov 11, 2026
Rank30ModelGOGemini 3.7 FlashGoogleAgent Score-7.35% (-8.80% - -5.90%)Percentile30.95%Participants43Observations303.5KEvidenceAEvaluatedNov 11, 2026

LM Arena Agent Steerability Highlights

The leading models and scores on this benchmark.

LM Arena Agent Steerability Score Distribution

A closer view of the leading scores on this benchmark.

LM Arena Agent Steerability

The Top AI Models for LM Arena Agent Steerability

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

What is LM Arena Agent Steerability?

What LM Arena Agent Steerability measures and how its scores work.

LM Arena Agent Steerability is a benchmark leaderboard.

It reports Agent Score as a ratio.

Scores are shown in ratio. This benchmark is verified and has an evidence level of A.

Family
lm-arena-agent-steerability
Modality
Unspecified
Primary category
Unspecified
Score direction
higher
LLMBoard eligible
No
Evaluation key
overall

LM Arena. Benchmark scores retain their original unit. Overall score eligibility is shown separately.

FAQ

Common questions about LM Arena Agent Steerability.

Which model scores highest on LM Arena Agent Steerability?

Gemini 4 Carbon is currently ranked first with 19.06%.

What are the top three models on LM Arena Agent Steerability?

The current leaders are Gemini 4 Carbon (19.06%), GPT-6 Astra (14.82%), and Claude Fable 5.1 (14.80%).

Which LM Arena Agent Steerability model has the lowest official input price?

Claude Haiku 5.5 has the lowest matched official input price at $0.10 input / $0.50 output per 1M tokens.

Which models are fastest among LM Arena Agent Steerability results?

The fastest matched records are MiniMax M3 (156.78 tok/s via MiniMax), Mistral Large 4 (107.24 tok/s via Mistral AI), and Gemini 3.1 Pro (90.00 tok/s via Google).

Does the highest agent 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 LM Arena Agent Steerability measure?

It reports Agent Score as a ratio.

Is a higher agent score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

43 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 #1Gemini 4 Carbon19.06% (18.45% - 19.68%)
Rank #2GPT-6 Astra14.82% (13.03% - 16.61%)
Rank #3Claude Fable 5.114.80% (12.96% - 16.64%)
Rank #4Gemini 4 Argon13.16% (12.18% - 14.13%)

Ranking basisThis lm arena agent steerability AI model leaderboard uses descending agent score in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.

  1. 01
    GO
    Gemini 4 CarbonGoogle
    Agent Score
    19.06%
    Price
    $2.0 input / $10 output per 1M tokens

    Strengths

    • Ranks #1 of 43 compared models
    • 100th percentile on this benchmark
    • A evidence result

    Considerations

    • This result measures LM Arena Agent Steerability, not total model capability
  2. 02
    OP
    GPT-6 AstraOpenAI
    Agent Score
    14.82%
    Price
    $10 input / $50 output per 1M tokens
    Speed
    Up to 86.85 tok/s via OpenAI

    Strengths

    • Ranks #2 of 43 compared models
    • 98th percentile on this benchmark
    • A evidence result

    Considerations

    • This result measures LM Arena Agent Steerability, not total model capability
  3. 03
    AN
    Anthropic
    Agent Score
    14.80%
    Price
    $10 input / $50 output per 1M tokens

    Strengths

    • Ranks #3 of 43 compared models
    • 95th percentile on this benchmark
    • A evidence result

    Considerations

    • This result measures LM Arena Agent Steerability, not total model capability
  4. 04
    GO
    Google
    Agent Score
    13.16%
    Price
    $2.0 input / $10 output per 1M tokens

    Strengths

    • Ranks #4 of 43 compared models
    • 93th percentile on this benchmark
    • A evidence result

    Considerations

    • This result measures LM Arena Agent Steerability, not total model capability
  5. 05
    TM
    Thinking Machines Lab
    Agent Score
    9.28%

    Strengths

    • Ranks #5 of 43 compared models
    • 90th percentile on this benchmark
    • A evidence result

    Considerations

    • This result measures LM Arena Agent Steerability, not total model capability

Selection summary

Best AI Models for LM Arena Agent Steerability

Gemini 4 Carbon currently leads LM Arena Agent Steerability with 19.06%. 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.

Claude Fable 5.1
Gemini 4 Argon
Inkling 2 Small
Benchmark rank #1Gemini 4 Carbon19.06% · $2.0 input / $10 output per 1M tokens
Benchmark rank #2GPT-6 Astra14.82% · $10 input / $50 output per 1M tokens
Benchmark rank #3Claude Fable 5.114.80% · $10 input / $50 output per 1M tokens