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

accuracy benchmark

Artificial Analysis Speech To Text Leaderboard

Artificial Analysis Speech To Text is an accuracy benchmark based on the official AA WER Index leaderboard published by Artificial Analysis.

Updated Oct 23, 2026

Models59
Model coverage59
MetricAA WER Index
EvidenceB

On this page

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

Artificial Analysis Speech To Text Ranking

Lower aa wer index ranks better on this benchmark.

30 of 59 rows
Columns

Show columns

Sort by
Rank
Model
AA WER Index
Percentile
Participants
Evidence
Evaluated
Rank01ModelALFun-Realtime-ASR-previewAlibabaAA WER Index0.02 indexPercentile100.00%Participants59EvidenceBEvaluatedN/A
Rank02ModelMIMAI-Transcribe-2MicrosoftAA WER Index0.02 indexPercentile98.28%Participants59EvidenceBEvaluatedN/A
Rank03ModelMAVoxtral Small Transcribe 3Mistral AIAA WER Index0.02 indexPercentile96.55%Participants59EvidenceBEvaluatedN/A
Rank04ModelELScribe v2ElevenLabsAA WER Index0.02 indexPercentile94.83%Participants59EvidenceBEvaluatedN/A
Rank05ModelMAVoxtral Mini Transcribe 3Mistral AIAA WER Index0.02 indexPercentile93.10%Participants59EvidenceBEvaluatedN/A
Rank06ModelXAGrok Voice Transcribe 2.0xAIAA WER Index0.02 indexPercentile91.38%Participants59EvidenceBEvaluatedN/A
Rank07ModelMIMAI-Transcribe-1.5MicrosoftAA WER Index0.02 indexPercentile89.66%Participants59EvidenceBEvaluatedN/A
Rank08ModelSTStepAudio 3 ASRStepFunAA WER Index0.03 indexPercentile87.93%Participants59EvidenceBEvaluatedN/A
Rank09ModelGOGemini 3.5 TranscribeGoogleAA WER Index0.03 indexPercentile86.21%Participants59EvidenceBEvaluatedN/A
Rank10ModelMAVoxtral SmallMistral AIAA WER Index0.03 indexPercentile84.48%Participants59EvidenceBEvaluatedN/A
Rank11ModelGOGemini 3.1 Pro Preview (High)GoogleAA WER Index0.03 indexPercentile82.76%Participants59EvidenceBEvaluatedN/A
Rank12ModelZOZoom Scribe V2 ProZoomAA WER Index0.03 indexPercentile81.03%Participants59EvidenceBEvaluatedN/A
Rank13ModelGOGemini 3 Flash (High)GoogleAA WER Index0.03 indexPercentile79.31%Participants59EvidenceBEvaluatedN/A
Rank14ModelGOGemini 2.5 ProGoogleAA WER Index0.03 indexPercentile77.59%Participants59EvidenceBEvaluatedN/A
Rank15ModelASUniversal-3.5 ProAssemblyAIAA WER Index0.03 indexPercentile75.86%Participants59EvidenceBEvaluatedN/A
Rank16ModelASUniversal-3 ProAssemblyAIAA WER Index0.03 indexPercentile74.14%Participants59EvidenceBEvaluatedN/A
Rank17ModelGLSolaria-3GladiaAA WER Index0.03 indexPercentile72.41%Participants59EvidenceBEvaluatedN/A
Rank18ModelOPGPT TranscribeOpenAIAA WER Index0.03 indexPercentile70.69%Participants59EvidenceBEvaluatedN/A
Rank19ModelSMSmallest AI PulseSmallest.aiAA WER Index0.03 indexPercentile68.97%Participants59EvidenceBEvaluatedN/A
Rank20ModelSMSmallest AI Pulse ProSmallest.aiAA WER Index0.03 indexPercentile67.24%Participants59EvidenceBEvaluatedN/A
Rank21ModelGOGemini 3.1 Flash-Lite Preview (Minimal)GoogleAA WER Index0.03 indexPercentile65.52%Participants59EvidenceBEvaluatedN/A
Rank22ModelREResonant-1Reson8AA WER Index0.03 indexPercentile63.79%Participants59EvidenceBEvaluatedN/A
Rank23ModelALQwen3.5 Omni PlusAlibabaAA WER Index0.04 indexPercentile62.07%Participants59EvidenceBEvaluatedN/A
Rank24ModelTMInkling (256K)Thinking Machines LabAA WER Index0.04 indexPercentile60.34%Participants59EvidenceBEvaluatedN/A
Rank25ModelGOGemini 3.1 Pro Preview (Low)GoogleAA WER Index0.04 indexPercentile58.62%Participants59EvidenceBEvaluatedN/A
Rank26ModelMAVoxtral Mini Transcribe 2Mistral AIAA WER Index0.04 indexPercentile56.90%Participants59EvidenceBEvaluatedN/A
Rank27ModelASUniversalAssemblyAIAA WER Index0.04 indexPercentile55.17%Participants59EvidenceBEvaluatedN/A
Rank28ModelSOSoniox v5 AsyncSonioxAA WER Index0.04 indexPercentile53.45%Participants59EvidenceBEvaluatedN/A
Rank29ModelINInworld STT 1InworldAA WER Index0.04 indexPercentile51.72%Participants59EvidenceBEvaluatedN/A
Rank30ModelOPGPT-4o TranscribeOpenAIAA WER Index0.04 indexPercentile50.00%Participants59EvidenceBEvaluatedN/A

Artificial Analysis Speech To Text Highlights

The leading models and scores on this benchmark.

Artificial Analysis Speech To Text Score Distribution

A closer view of the leading scores on this benchmark.

Artificial Analysis Speech To Text

The Top AI Models for Artificial Analysis Speech To Text

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

What is Artificial Analysis Speech To Text?

What Artificial Analysis Speech To Text measures and how its scores work.

Artificial Analysis Speech To Text is a benchmark for speech-to-text systems.

It reports the AA WER Index in index units.

Scores are shown in index. This benchmark is verified and has an evidence level of B.

Family
Artificial Analysis Speech To Text
Modality
audio
Primary category
accuracy
Score direction
lower
LLMBoard eligible
No
Evaluation key
overall

Artificial Analysis. Benchmark scores retain their original unit. Overall score eligibility is shown separately.

FAQ

Common questions about Artificial Analysis Speech To Text.

Which model scores highest on Artificial Analysis Speech To Text?

Fun-Realtime-ASR-preview is currently ranked first with 0.02 index.

What are the top three models on Artificial Analysis Speech To Text?

The current leaders are Fun-Realtime-ASR-preview (0.02 index), MAI-Transcribe-2 (0.02 index), and Voxtral Small Transcribe 3 (0.02 index).

Which Artificial Analysis Speech To Text model has the lowest official input price?

No matched official input price is currently available.

Which models are fastest among Artificial Analysis Speech To Text results?

No matched runtime record is currently available.

Does the highest aa wer index 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 Artificial Analysis Speech To Text measure?

It reports the AA WER Index in index units.

Is a higher aa wer index better?

No. Lower values rank better for this benchmark.

How many models are compared?

59 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 #1Fun-Realtime-ASR-preview0.02 index
Rank #2MAI-Transcribe-20.02 index
Rank #3Voxtral Small Transcribe 30.02 index
Rank #4Scribe v20.02 index

Ranking basisThis artificial analysis speech to text AI model leaderboard uses ascending aa wer index in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.

  1. 01
    AL
    Fun-Realtime-ASR-previewAlibaba
    AA WER Index
    0.02 index

    Strengths

    • Ranks #1 of 59 compared models
    • 100th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures Artificial Analysis Speech To Text, not total model capability
  2. 02
    MI
    MAI-Transcribe-2Microsoft
    AA WER Index
    0.02 index

    Strengths

    • Ranks #2 of 59 compared models
    • 98th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures Artificial Analysis Speech To Text, not total model capability
  3. 03
    MA
    Mistral AI
    AA WER Index
    0.02 index

    Strengths

    • Ranks #3 of 59 compared models
    • 97th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures Artificial Analysis Speech To Text, not total model capability
  4. 04
    EL
    ElevenLabs
    AA WER Index
    0.02 index

    Strengths

    • Ranks #4 of 59 compared models
    • 95th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures Artificial Analysis Speech To Text, not total model capability
  5. 05
    MA
    Mistral AI
    AA WER Index
    0.02 index

    Strengths

    • Ranks #5 of 59 compared models
    • 93th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures Artificial Analysis Speech To Text, not total model capability

Selection summary

Best AI Models for Artificial Analysis Speech To Text

Fun-Realtime-ASR-preview currently leads Artificial Analysis Speech To Text with 0.02 index. 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.

Voxtral Small Transcribe 3
Scribe v2
Voxtral Mini Transcribe 3
Benchmark rank #1Fun-Realtime-ASR-preview0.02 index
Benchmark rank #2MAI-Transcribe-20.02 index
Benchmark rank #3Voxtral Small Transcribe 30.02 index