Benchmark Almanac

OpenAudioBench Llama Questions

GPT-4o-Audio 88.4% tops its leaderboard Baichuan-Omni-1.5 78.5% leads the models we track
Not in model cards

Models answer spoken Llama Questions. The score is the published accuracy.

One track of OpenAudioBench. The OpenAudioBench record keeps lab scores that combine tracks.

Made by
Baichuan
Released
January 2025
Kind
Audio understanding and voice
Level
Everyday
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
Last model added
January 2025
Scores refreshed
November 12th, 2026
5
Entries on its own leaderboard, 4 models we track

GPT-4o-Audio tops the leaderboard with 88.4%

Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Baichuan-Omni-1.5, at 78.5%.

The newest model with a score here came out in January 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 88.4 GPT-4o-Audio
  2. 78.5 Baichuan-Omni-1.5
  3. 77.8 MiniCPM-o-2_6
  4. 74.2 VITA-1.5
  5. 71.0 GLM-4-Voice-9B
  1. 1 GPT-4o-Audio speech-to-text 88.4%
  2. 2 Baichuan-Omni-1.5 speech-to-text 78.5%
  3. 3 MiniCPM-o-2_6 speech-to-text 77.8%
  4. 4 VITA-1.5 speech-to-text 74.2%
  5. 5 GLM-4-Voice-9B speech-to-speech 71%

Source: Official OpenAudioBench Llama Questions leaderboard

#EntryScoreSettingDateSource
1 GPT-4o-Audio 88.4% speech-to-text Official leaderboard
2 Baichuan-Omni-1.5 78.5% speech-to-text Jan 2025 Official leaderboard
3 MiniCPM-o-2_6 77.8% speech-to-text Jan 2025 Official leaderboard
4 VITA-1.5 74.2% speech-to-text Dec 2024 Official leaderboard
5 GLM-4-Voice-9B 71% speech-to-speech Oct 2024 Official leaderboard

3 models have held number one on the official leaderboard since its release in January 2025

Lab claims count from the day the card came out, independent runs from the model’s release.

Other labs Baichuan
Own leaderboard Not tested since Jan 2025
Baichuan-Omni-1.5 78.5%

Source: Official OpenAudioBench Llama Questions leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Baichuan-Omni-1.5 VITA-1.5 GLM-4-Voice-9B 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Baichuan-Omni-1.5 GLM-4-Voice-9B

Source: Official OpenAudioBench Llama Questions leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Audio and video

All 233 in Audio and video
  • Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy. Qwen 3.8 Max 88.7%
  • LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization. Gemini 4 Argon 91.7%
  • MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy. Qwen 3.8 Max 82.4%
  • MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy. Qwen 3.8 Max 90.8%
  • FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better. Phi-4-multimodal 4.00%
  • VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules. Qwen3.5-Omni-Plus 93.1%