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
1224 models729 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

agents benchmark

DeepSWE 1.1 Leaderboard

DeepSWE 1.1 evaluates software engineering agents using the mini-swe-agent harness.

Updated Sep 5, 2026

Models31
Model coverage31
MetricScore
EvidenceB

On this page

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

DeepSWE 1.1 Ranking

Higher score ranks better on this benchmark.

30 of 31 rows
Columns

Show columns

Sort by
Rank
Model
Score
Percentile
Participants
Evidence
Evaluated
Rank01ModelMEMuse Spark 1.3MetaScore75.40%Percentile100.00%Participants31EvidenceCEvaluatedSep 8, 2026
Rank02ModelOPGPT-6 AstraOpenAIScore74.10%Percentile96.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank03ModelGOGemini 3.8 FlashGoogleScore73.70%Percentile93.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank04ModelOPGPT-5.6 SolOpenAIScore73.00%Percentile90.00%Participants31EvidenceBEvaluatedSep 8, 2026
Rank05ModelANClaude Fable 5AnthropicScore70.00%Percentile86.67%Participants31EvidenceBEvaluatedSep 8, 2026
Rank06ModelOPGPT-5.6 TerraOpenAIScore70.00%Percentile83.33%Participants31EvidenceBEvaluatedSep 8, 2026
Rank07ModelMAKimi K3Moonshot AIScore69.00%Percentile80.00%Participants31EvidenceBEvaluatedSep 8, 2026
Rank08ModelANClaude Opus 5AnthropicScore68.80%Percentile76.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank09ModelOPGPT-5.5OpenAIScore67.00%Percentile73.33%Participants31EvidenceBEvaluatedSep 8, 2026
Rank10ModelOPGPT-5.6 LunaOpenAIScore67.00%Percentile70.00%Participants31EvidenceBEvaluatedSep 8, 2026
Rank11ModelZAGLM-5.3Zhipu AIScore66.90%Percentile66.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank12ModelXAGrok 4.6xAIScore65.90%Percentile63.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank13ModelGOGemini 3.7 FlashGoogleScore65.30%Percentile60.00%Participants31EvidenceCEvaluatedSep 8, 2026
Rank14ModelZAGLM-5.3-FlashZhipu AIScore63.40%Percentile56.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank15ModelMEMuse Spark 1.2MetaScore59.30%Percentile53.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank16ModelANClaude Opus 4.8AnthropicScore59.00%Percentile50.00%Participants31EvidenceBEvaluatedSep 8, 2026
Rank17ModelACQwen3.8-Flash-NextAlibaba Cloud / Qwen TeamScore58.70%Percentile46.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank18ModelACQwen3.8 FlashAlibaba Cloud / Qwen TeamScore58.70%Percentile43.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank19ModelACQwen3.8 MaxAlibaba Cloud / Qwen TeamScore56.60%Percentile40.00%Participants31EvidenceCEvaluatedSep 8, 2026
Rank20ModelANClaude Sonnet 5AnthropicScore54.00%Percentile36.67%Participants31EvidenceBEvaluatedSep 8, 2026
Rank21ModelXAGrok 4.5xAIScore54.00%Percentile33.33%Participants31EvidenceBEvaluatedSep 8, 2026
Rank22ModelMEMuse Spark 1.1MetaScore53.00%Percentile30.00%Participants31EvidenceBEvaluatedSep 8, 2026
Rank23ModelOPGPT-5.4OpenAIScore52.00%Percentile26.67%Participants31EvidenceBEvaluatedSep 8, 2026
Rank24ModelGOGemini 3.6 FlashGoogleScore49.00%Percentile23.33%Participants31EvidenceBEvaluatedSep 8, 2026
Rank25ModelZAGLM-5.2Zhipu AIScore44.00%Percentile20.00%Participants31EvidenceBEvaluatedSep 8, 2026
Rank26ModelACQwen3.8-27BAlibaba Cloud / Qwen TeamScore42.20%Percentile16.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank27ModelPOLaguna S 2.1PoolsideScore40.40%Percentile13.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank28ModelGOGemini 3.5 FlashGoogleScore37.00%Percentile10.00%Participants31EvidenceBEvaluatedSep 8, 2026
Rank29ModelMAKimi K2.7 CodeMoonshot AIScore31.00%Percentile6.67%Participants31EvidenceBEvaluatedSep 8, 2026
Rank30ModelANClaude Sonnet 4.6AnthropicScore30.00%Percentile3.33%Participants31EvidenceBEvaluatedSep 8, 2026

DeepSWE 1.1 Highlights

The leading models and scores on this benchmark.

DeepSWE 1.1 Score Distribution

A closer view of the leading scores on this benchmark.

DeepSWE 1.1

The Top AI Models for DeepSWE 1.1

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

What is DeepSWE 1.1?

What DeepSWE 1.1 measures and how its scores work.

DeepSWE 1.1 is a benchmark for evaluating software engineering agents.

It evaluates software engineering agents.

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

Family
DeepSWE 1.1
Modality
text
Primary category
agents
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 DeepSWE 1.1.

Which model scores highest on DeepSWE 1.1?

Muse Spark 1.3 is currently ranked first with 75.40%.

What are the top three models on DeepSWE 1.1?

The current leaders are Muse Spark 1.3 (75.40%), GPT-6 Astra (74.10%), and Gemini 3.8 Flash (73.70%).

Which DeepSWE 1.1 model has the lowest official input price?

GLM-5.3-Flash has the lowest matched official input price at $0.08 input / $0.25 output per 1M tokens.

Which models are fastest among DeepSWE 1.1 results?

The fastest matched records are GLM-5.3 (472.69 tok/s via FriendliAI), Gemini 3.5 Flash (210.94 tok/s via Google), and GPT-5.5 (134.94 tok/s via OpenAI).

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 DeepSWE 1.1 measure?

It evaluates software engineering agents.

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

31 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 #1Muse Spark 1.375.40%
Rank #2GPT-6 Astra74.10%
Rank #3Gemini 3.8 Flash73.70%
Rank #4GPT-5.6 Sol73.00%

Ranking basisThis deepswe 1.1 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
    ME
    Muse Spark 1.3Meta
    Score
    75.40%
    Price
    $1.3 input / $4.3 output per 1M tokens
    Speed
    Up to 8.44 tok/s via Meta Model API

    Strengths

    • Ranks #1 of 31 compared models
    • 100th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures DeepSWE 1.1, not total model capability
  2. 02
    OP
    GPT-6 AstraOpenAI
    Score
    74.10%
    Price
    $10 input / $50 output per 1M tokens
    Speed
    Up to 30.66 tok/s via OpenAI

    Strengths

    • Ranks #2 of 31 compared models
    • 97th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures DeepSWE 1.1, not total model capability
  3. 03
    GO
    Google
    Score
    73.70%
    Price
    $0.75 input / $3.8 output per 1M tokens
    Speed
    Up to 51.56 tok/s via Google

    Strengths

    • Ranks #3 of 31 compared models
    • 93th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures DeepSWE 1.1, not total model capability
  4. 04
    OP
    OpenAI
    Score
    73.00%
    Price
    $4.0 input / $20 output per 1M tokens
    Speed
    Up to 2.36 tok/s via OpenAI

    Strengths

    • Ranks #4 of 31 compared models
    • 90th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures DeepSWE 1.1, not total model capability
  5. 05
    AN
    Anthropic
    Score
    70.00%
    Price
    $10 input / $50 output per 1M tokens
    Speed
    Up to 43.18 tok/s via Anthropic

    Strengths

    • Ranks #5 of 31 compared models
    • 87th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures DeepSWE 1.1, not total model capability

Selection summary

Best AI Models for DeepSWE 1.1

Muse Spark 1.3 currently leads DeepSWE 1.1 with 75.40%. 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
Claude Fable 5
Benchmark rank #1Muse Spark 1.375.40% · $1.3 input / $4.3 output per 1M tokens
Benchmark rank #2GPT-6 Astra74.10% · $10 input / $50 output per 1M tokens
Benchmark rank #3Gemini 3.8 Flash73.70% · $0.75 input / $3.8 output per 1M tokens