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
1450 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

agents benchmark

NL2Repo Leaderboard

NL2Repo evaluates long-horizon coding capabilities, including repository-level understanding, by requiring models to generate or modify code across entire repositories from natural language specifications.

Updated Nov 17, 2026

Models27
Model coverage27
MetricScore
EvidenceB

On this page

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

NL2Repo Ranking

Higher score ranks better on this benchmark.

27 rows
Columns

Show columns

Sort by
Rank
Model
Score
Percentile
Participants
Evidence
Evaluated
Rank01ModelDEDeepSeek-V4.1-Flash-1017DeepSeekScore68.20%Percentile100.00%Participants27EvidenceCEvaluatedNov 17, 2026
Rank02ModelBAERNIE 5.5 Flash PreviewBaiduScore64.80%Percentile96.15%Participants27EvidenceCEvaluatedNov 17, 2026
Rank03ModelDEDeepSeek-V4.1-FlashDeepSeekScore64.00%Percentile92.31%Participants27EvidenceCEvaluatedNov 17, 2026
Rank04ModelDEDeepSeek-V4-Pro-0813DeepSeekScore61.50%Percentile88.46%Participants27EvidenceCEvaluatedNov 17, 2026
Rank05ModelMIMiniMax M3.1 FlashMiniMaxScore61.20%Percentile84.62%Participants27EvidenceCEvaluatedNov 17, 2026
Rank06ModelMELongCat-2.5MeituanScore60.50%Percentile80.77%Participants27EvidenceCEvaluatedNov 17, 2026
Rank07ModelTEHy4 previewTencentScore58.90%Percentile76.92%Participants27EvidenceCEvaluatedNov 17, 2026
Rank08ModelZAGLM-5.3Zhipu AIScore58.00%Percentile73.08%Participants27EvidenceCEvaluatedNov 17, 2026
Rank09ModelDEDeepSeek-V4-Flash-Vision-ExpDeepSeekScore57.70%Percentile69.23%Participants27EvidenceCEvaluatedNov 17, 2026
Rank10ModelZAGLM-5.3-FlashZhipu AIScore56.30%Percentile65.38%Participants27EvidenceCEvaluatedNov 17, 2026
Rank11ModelACQwen3.8 MaxAlibaba Cloud / Qwen TeamScore55.90%Percentile61.54%Participants27EvidenceCEvaluatedNov 17, 2026
Rank12ModelDEDeepSeek-V4-Flash-0731DeepSeekScore54.20%Percentile57.69%Participants27EvidenceCEvaluatedNov 17, 2026
Rank13ModelZAGLM-5.2Zhipu AIScore48.90%Percentile53.85%Participants27EvidenceCEvaluatedNov 17, 2026
Rank14ModelACQwen3.8-Flash-NextAlibaba Cloud / Qwen TeamScore48.10%Percentile50.00%Participants27EvidenceCEvaluatedNov 17, 2026
Rank15ModelACQwen3.8 FlashAlibaba Cloud / Qwen TeamScore48.10%Percentile46.15%Participants27EvidenceCEvaluatedNov 17, 2026
Rank16ModelACQwen3.7 MaxAlibaba Cloud / Qwen TeamScore47.20%Percentile42.31%Participants27EvidenceCEvaluatedNov 17, 2026
Rank17ModelBYSeed 2.1 ProByteDanceScore47.00%Percentile38.46%Participants27EvidenceCEvaluatedNov 17, 2026
Rank18ModelTEHy3TencentScore45.60%Percentile34.62%Participants27EvidenceCEvaluatedNov 17, 2026
Rank19ModelBYSeed 2.1 TurboByteDanceScore43.70%Percentile30.77%Participants27EvidenceCEvaluatedNov 17, 2026
Rank20ModelZAGLM-5.1Zhipu AIScore42.70%Percentile26.92%Participants27EvidenceCEvaluatedNov 17, 2026
Rank21ModelACQwen3.8-27BAlibaba Cloud / Qwen TeamScore42.30%Percentile23.08%Participants27EvidenceCEvaluatedNov 17, 2026
Rank22ModelMIMiniMax M3MiniMaxScore42.13%Percentile19.23%Participants27EvidenceCEvaluatedNov 17, 2026
Rank23ModelACQwen3.7-PlusAlibaba Cloud / Qwen TeamScore41.10%Percentile15.38%Participants27EvidenceCEvaluatedNov 17, 2026
Rank24ModelMIMiniMax M2.7MiniMaxScore39.80%Percentile11.54%Participants27EvidenceCEvaluatedNov 17, 2026
Rank25ModelACQwen3.6 PlusAlibaba Cloud / Qwen TeamScore37.90%Percentile7.69%Participants27EvidenceCEvaluatedNov 17, 2026
Rank26ModelACQwen3.6-27BAlibaba Cloud / Qwen TeamScore36.20%Percentile3.85%Participants27EvidenceCEvaluatedNov 17, 2026
Rank27ModelACQwen3.6-35B-A3BAlibaba Cloud / Qwen TeamScore29.40%Percentile0.00%Participants27EvidenceCEvaluatedNov 17, 2026

NL2Repo Highlights

The leading models and scores on this benchmark.

NL2Repo Score Distribution

A closer view of the leading scores on this benchmark.

NL2Repo

The Top AI Models for NL2Repo

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

What is NL2Repo?

What NL2Repo measures and how its scores work.

NL2Repo is an agents benchmark for evaluating coding capabilities across entire repositories from natural language specifications.

It measures long-horizon coding capabilities and repository-level understanding using the Score metric, reported as a ratio.

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

Family
NL2Repo
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 NL2Repo.

Which model scores highest on NL2Repo?

DeepSeek-V4.1-Flash-1017 is currently ranked first with 68.20%.

What are the top three models on NL2Repo?

The current leaders are DeepSeek-V4.1-Flash-1017 (68.20%), ERNIE 5.5 Flash Preview (64.80%), and DeepSeek-V4.1-Flash (64.00%).

Which NL2Repo 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 NL2Repo results?

The fastest matched records are MiniMax M3.1 Flash (74.86 tok/s via MiniMax), GLM-5.3-Flash (63.07 tok/s via FriendliAI), and MiniMax M3 (36.52 tok/s via MiniMax).

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 NL2Repo measure?

It measures long-horizon coding capabilities and repository-level understanding using the Score metric, reported as a ratio.

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

27 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-Flash-101768.20%
Rank #2ERNIE 5.5 Flash Preview64.80%
Rank #3DeepSeek-V4.1-Flash64.00%
Rank #4DeepSeek-V4-Pro-081361.50%

Ranking basisThis nl2repo 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-Flash-1017DeepSeek
    Score
    68.20%
    Price
    $0.15 input / $0.60 output per 1M tokens

    Strengths

    • Ranks #1 of 27 compared models
    • 100th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures NL2Repo, not total model capability
  2. 02
    BA
    ERNIE 5.5 Flash PreviewBaidu
    Score
    64.80%

    Strengths

    • Ranks #2 of 27 compared models
    • 96th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures NL2Repo, not total model capability
  3. 03
    DE
    DeepSeek
    Score
    64.00%
    Price
    $0.15 input / $0.60 output per 1M tokens

    Strengths

    • Ranks #3 of 27 compared models
    • 92th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures NL2Repo, not total model capability
  4. 04
    DE
    DeepSeek
    Score
    61.50%
    Price
    $0.66 input / $2.0 output per 1M tokens

    Strengths

    • Ranks #4 of 27 compared models
    • 88th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures NL2Repo, not total model capability
  5. 05
    MI
    MiniMax
    Score
    61.20%
    Price
    $0.15 input / $0.60 output per 1M tokens
    Speed
    Up to 74.86 tok/s via MiniMax

    Strengths

    • Ranks #5 of 27 compared models
    • 85th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures NL2Repo, not total model capability

Selection summary

Best AI Models for NL2Repo

DeepSeek-V4.1-Flash-1017 currently leads NL2Repo with 68.20%. 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.

DeepSeek-V4.1-Flash
DeepSeek-V4-Pro-0813
MiniMax M3.1 Flash
Benchmark rank #1DeepSeek-V4.1-Flash-101768.20% · $0.15 input / $0.60 output per 1M tokens
Benchmark rank #2ERNIE 5.5 Flash Preview64.80%
Benchmark rank #3DeepSeek-V4.1-Flash64.00% · $0.15 input / $0.60 output per 1M tokens