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

math benchmark

MGSM Leaderboard

MGSM is a benchmark of 250 grade-school math problems manually translated from the GSM8K dataset into ten typologically diverse languages.

Updated Sep 5, 2026

Models31
Model coverage31
MetricScore
EvidenceB

On this page

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

MGSM Ranking

Higher score ranks better on this benchmark.

30 of 31 rows
Columns

Show columns

Sort by
Rank
Model
Score
Percentile
Participants
Evidence
Evaluated
Rank01ModelMELlama 4 MaverickMetaScore92.30%Percentile100.00%Participants31EvidenceCEvaluatedSep 8, 2026
Rank02ModelOPo3-miniOpenAIScore92.00%Percentile96.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank03ModelANClaude 3.5 SonnetAnthropicScore91.60%Percentile93.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank04ModelANClaude 3.5 SonnetAnthropicScore91.60%Percentile90.00%Participants31EvidenceCEvaluatedSep 8, 2026
Rank05ModelMELlama 3.3 70B InstructMetaScore91.10%Percentile86.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank06ModelOPo1-previewOpenAIScore90.80%Percentile83.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank07ModelANClaude 3 OpusAnthropicScore90.70%Percentile80.00%Participants31EvidenceCEvaluatedSep 8, 2026
Rank08ModelMELlama 4 ScoutMetaScore90.60%Percentile76.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank09ModelOPGPT-4oOpenAIScore90.50%Percentile73.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank10ModelOPo1OpenAIScore89.30%Percentile70.00%Participants31EvidenceCEvaluatedSep 8, 2026
Rank11ModelOPGPT-4 TurboOpenAIScore88.50%Percentile66.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank12ModelGOGemini 1.5 ProGoogleScore87.50%Percentile63.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank13ModelOPGPT-4o miniOpenAIScore87.00%Percentile60.00%Participants31EvidenceCEvaluatedSep 8, 2026
Rank14ModelMELlama 3.2 90B InstructMetaScore86.90%Percentile56.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank15ModelANClaude 3.5 HaikuAnthropicScore85.60%Percentile53.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank16ModelACQwen3 235B A22BAlibaba Cloud / Qwen TeamScore83.53%Percentile50.00%Participants31EvidenceCEvaluatedSep 8, 2026
Rank17ModelANClaude 3 SonnetAnthropicScore83.50%Percentile46.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank18ModelGOGemini 1.5 FlashGoogleScore82.60%Percentile43.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank19ModelMIPhi 4MicrosoftScore80.60%Percentile40.00%Participants31EvidenceCEvaluatedSep 8, 2026
Rank20ModelANClaude 3 HaikuAnthropicScore75.10%Percentile36.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank21ModelOPGPT-4OpenAIScore74.50%Percentile33.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank22ModelMELlama 3.2 11B InstructMetaScore68.90%Percentile30.00%Participants31EvidenceCEvaluatedSep 8, 2026
Rank23ModelGOGemma 3n E4B InstructedGoogleScore67.00%Percentile26.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank24ModelMIPhi 4 MiniMicrosoftScore63.90%Percentile23.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank25ModelGOGemma 3n E4B Instructed LiteRT PreviewGoogleScore60.70%Percentile20.00%Participants31EvidenceCEvaluatedSep 8, 2026
Rank26ModelMIPhi-3.5-MoE-instructMicrosoftScore58.70%Percentile16.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank27ModelMELlama 3.2 3B InstructMetaScore58.20%Percentile13.33%Participants31EvidenceCEvaluatedSep 8, 2026
Rank28ModelOPGPT-3.5 TurboOpenAIScore56.30%Percentile10.00%Participants31EvidenceBEvaluatedSep 8, 2026
Rank29ModelGOGemma 3n E2B InstructedGoogleScore53.10%Percentile6.67%Participants31EvidenceCEvaluatedSep 8, 2026
Rank30ModelGOGemma 3n E2B Instructed LiteRT (Preview)GoogleScore53.10%Percentile3.33%Participants31EvidenceCEvaluatedSep 8, 2026

MGSM Highlights

The leading models and scores on this benchmark.

MGSM Score Distribution

A closer view of the leading scores on this benchmark.

MGSM

The Top AI Models for MGSM

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

What is MGSM?

What MGSM measures and how its scores work.

MGSM (Multilingual Grade School Math) is a benchmark containing grade-school math problems in Spanish, French, German, Russian, Chinese, Japanese, Thai, Swahili, Bengali, and Telugu.

MGSM evaluates multilingual mathematical reasoning capabilities 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
MGSM
Modality
text
Primary category
math
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 MGSM.

Which model scores highest on MGSM?

Llama 4 Maverick is currently ranked first with 92.30%.

What are the top three models on MGSM?

The current leaders are Llama 4 Maverick (92.30%), o3-mini (92.00%), and Claude 3.5 Sonnet (91.60%).

Which MGSM model has the lowest official input price?

Phi 4 Mini has the lowest matched official input price at $0.08 input / $0.30 output per 1M tokens.

Which models are fastest among MGSM results?

The fastest matched records are Llama 3.3 70B Instruct (2,220.00 tok/s via Cerebras), Llama 4 Scout (776.10 tok/s via Groq), and Llama 4 Maverick (307.30 tok/s via Groq).

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 MGSM measure?

MGSM evaluates multilingual mathematical reasoning capabilities 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?

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 #1Llama 4 Maverick92.30%
Rank #2o3-mini92.00%
Rank #3Claude 3.5 Sonnet91.60%
Rank #4Claude 3.5 Sonnet91.60%

Ranking basisThis mgsm 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
    Llama 4 MaverickMeta
    Score
    92.30%
    Speed
    Up to 307.30 tok/s via Groq

    Strengths

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

    Considerations

    • This result measures MGSM, not total model capability
  2. 02
    OP
    o3-miniOpenAI
    Score
    92.00%
    Price
    $1.1 input / $4.4 output per 1M tokens
    Speed
    Up to 115.00 tok/s via Azure

    Strengths

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

    Considerations

    • This result measures MGSM, not total model capability
  3. 03
    AN
    Anthropic
    Score
    91.60%
    Speed
    Up to 42.00 tok/s via Google

    Strengths

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

    Considerations

    • This result measures MGSM, not total model capability
  4. 04
    AN
    Anthropic
    Score
    91.60%
    Speed
    Up to 100.00 tok/s via Anthropic

    Strengths

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

    Considerations

    • This result measures MGSM, not total model capability
  5. 05
    ME
    Meta
    Score
    91.10%
    Speed
    Up to 2,220.00 tok/s via Cerebras

    Strengths

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

    Considerations

    • This result measures MGSM, not total model capability

Selection summary

Best AI Models for MGSM

Llama 4 Maverick currently leads MGSM with 92.30%. 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 3.5 Sonnet
Claude 3.5 Sonnet
Llama 3.3 70B Instruct
Benchmark rank #1Llama 4 Maverick92.30% · Up to 307.30 tok/s via Groq
Benchmark rank #2o3-mini92.00% · $1.1 input / $4.4 output per 1M tokens
Benchmark rank #3Claude 3.5 Sonnet91.60% · Up to 42.00 tok/s via Google