IBM Granite: API 제공업체 벤치마킹 및 분석
Artificial Analysis 단어 오류율 지수, 속도, 가격을 포함한 성능 지표에서 IBM Granite API 제공업체 분석.
주요 내용
API별 Artificial Analysis 단어 오류율(AA-WER) 지수
API별 Artificial Analysis 단어 오류율(AA-WER) 지수
% of words transcribed incorrectly · Lower is better · AA-WER v2 incorporates 3 datasets: AA-AgentTalk (50%), VoxPopuli-Cleaned-AA (25%), Earnings22-Cleaned-AA (25%)
참고: Earnings22에서 시간 제한으로 인해 모델이 전체 길이의 오디오를 안정적으로 처리할 수 없는 경우 약 9분 단위로 분할합니다(해당 모델: Nova 2 Pro, Amazon GPT-4o Transcribe, OpenAI GPT-4o Mini Transcribe, OpenAI). 시간 제한이 더 짧은 모델은 약 30초 단위로 분할합니다(해당 모델: Canary Qwen 2.5B, NVIDIA Inworld STT 1).
API 벤치마크
Artificial Analysis 단어 오류율 지수 vs. 가격
% of words transcribed incorrectly · Lower is better · AA-WER v2 incorporates 3 datasets: AA-AgentTalk (50%), VoxPopuli-Cleaned-AA (25%), Earnings22-Cleaned-AA (25%) · USD per 1000 minutes of audio
Most attractive quadrant
속도 계수
속도 계수
Input audio seconds transcribed per second · Higher is better
가격
전사 가격
USD per 1000 minutes of audio
주요 지표 요약 및 추가 정보
제공업체 | 추가 세부 정보 | ||||
|---|---|---|---|---|---|
Qwen3.5 Omni Flash |  Alibaba Cloud | 13.5% | 75.3 | 0.00 | |
Qwen3.5 Omni Plus | Alibaba Cloud | 3.5% | 95.8 | 0.00 | |
Fun-Realtime-ASR-preview | Alibaba Cloud | 1.7% | 20.5 | 0.00 | |
Nova 2 Pro |  Amazon Bedrock | 4.9% | 22.8 | 3.10 | |
Amazon Transcribe | Amazon Bedrock | 4.1% | 32.6 | 6.00 | |
Universal-3 Pro |  AssemblyAI | 3.1% | 83.4 | 3.50 | |
Universal, AssemblyAI | AssemblyAI | 3.8% | 116.1 | 2.50 | |
MAI-Transcribe-2 |  Microsoft AI | 2.0% | 290.1 | 1.67 | |
MAI-Transcribe-1.5 | Microsoft AI | 2.4% | 192.9 | 6.00 | |
MAI-Transcribe-1 | Microsoft AI | 2.6% | 68.3 | 6.00 | |
transcribe-03-2026 |  Cohere | 4.6% | 107.9 | 0.00 | |
Nova-3 |  Deepgram | 5.2% | 541.9 | 4.30 | |
Nova-3, Telnyx |  Telnyx | 4.8% | 417.6 | 7.40 | |
Nova-2, Telnyx | Telnyx | 5.1% | 459.4 | 7.40 | |
Scribe v2 |  ElevenLabs | 2.2% | 60.4 | 3.67 | |
Solaria-1, Gladia |  Gladia | 4.1% | 79.9 | 10.17 | |
Solaria-3, Gladia | Gladia | 3.2% | 65.2 | 10.16 | |
Chirp 3, Google |  Google | 4.3% | 30.2 | 16.00 | |
Chirp | Google | 31.2% | 14.1 | 16.00 | |
Gemini 3.5 Transcribe | Google | 2.6% | 87.6 | 5.00 | |
Gemini 3.1 Pro Preview (High) | Google | 2.8% | 7.2 | 18.15 | |
Gemini 3.1 Pro Preview (Low) | Google | 3.6% | 6.3 | 7.72 | |
Gemini 3 Flash (High) | Google | 2.9% | 19.0 | 13.70 | |
Gemini 2.5 Flash Lite | Google | 5.2% | 83.7 | 6.56 | |
Gemini 2.5 Flash | Google | 5.1% | 73.9 | 6.66 | |
Gemini 2.5 Pro | Google | 2.9% | 12.2 | 11.39 | |
Gemini 3.1 Flash-Lite Preview (Minimal) | Google | 3.4% | 82.3 | 5.83 | |
Gradium Speech-to-Text |  Gradium | 6.8% | 2.3 | 13.00 | |
Grok Speech to Text, SpaceXAI |  SpaceXAI | 4.0% | 234.6 | 1.67 | |
Inworld STT 1 |  Inworld | 3.9% | 157.3 | 2.50 | |
Voxtral Mini Transcribe 2 |  Mistral | 3.6% | 84.1 | 3.00 | |
Voxtral Small | Mistral | 2.8% | 66.7 | 4.00 | |
Voxtral Mini |  DeepInfra | 3.8% | 78.3 | 1.00 | |
Modulate STT Batch English VFast |  Modulate | 4.2% | 53.5 | 0.42 | |
Canary Qwen 2.5B, NVIDIA |  Replicate | 4.3% | 5.8 | 0.74 | |
Parakeet TDT 0.6B V2, NVIDIA |  NVIDIA | 6.4% | 95.9 | 0.00 | |
Parakeet RNNT 1.1B | Replicate | 5.4% | 6.3 | 1.91 | |
GPT Transcribe, OpenAI |  OpenAI | 3.3% | 38.1 | 4.50 | |
GPT-4o Transcribe | OpenAI | 4.0% | 36.2 | 6.00 | |
GPT-4o Mini Transcribe | OpenAI | 4.5% | 43.1 | 3.00 | |
Smallest AI Pulse Pro |  Smallest.ai | 2.4% | 273.8 | 4.00 | |
Resonant-1 |  Reson8 | 3.4% | 331.7 | 3.60 | |
Rev AI |  Rev AI | 5.9% | 13.0 | 3.33 | |
Smallest AI Pulse | Smallest.ai | 4.4% | 289.1 | 5.00 | |
Soniox v5 Async |  Soniox | 3.8% | 37.1 | 1.66 | |
Soniox V4 | Soniox | 3.9% | 44.3 | 1.66 | |
Speechmatics Melia |  Speechmatics | 4.9% | 178.2 | 4.00 | |
Speechmatics Standard | Speechmatics | 5.1% | 99.7 | 7.50 | |
Speechmatics Enhanced | Speechmatics | 4.0% | 69.8 | 12.50 | |
StepAudio 2.5 ASR, StepFun |  StepFun | 4.7% | 91.1 | 0.37 | |
StepAudio 3 ASR, StepFun | StepFun | 1.7% | 83.7 | 7.00 | |
Whisper Large v3 Turbo |  Groq | 4.6% | 145.8 | 0.67 | |
Whisper Large v3 Turbo, Telnyx | Telnyx | 5.5% | 30.5 | 15.00 | |
Wizper Large v3 |  fal.ai | 4.7% | 287.5 | 0.50 | |
Incredibly Fast Whisper | Replicate | 5.7% | 56.1 | 1.49 | |
Whisper Large v3 | Replicate | 10.1% | 3.0 | 4.23 | |
Whisper Large v3 | fal.ai | 4.1% | 78.7 | 1.15 | |
Whisper Large v2 | OpenAI | 4.1% | 29.0 | 6.00 |