AudioMC (text output) leaderboard
As of 6 Nov 2026, the top score on AudioMC (text output) is 60.4% by Gemini 3.8 Flash (Google), ahead of Salesforce Launa Omni 1 at 59.7%, across 21 model configurations with a published result (Scale AI SEAL, result dated 16 Oct 2026). The AudioMC questions with the model answering in text. AudioMC (text output) is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 6 Nov 2026)
- Gemini 3.8 Flash holds the top AudioMC (text output) score at 60.4% (Scale AI SEAL, 16 Oct 2026).
- Salesforce Launa Omni 1 is second on AudioMC (text output) at 59.7% (Scale AI SEAL, 16 Oct 2026).
- AudioMC (text output): AudioMC answered in text. Scale AI (Scale AI SEAL).
- 21 model configurations have a published AudioMC (text output) result on BenchLeader (BenchLeader, 6 Nov 2026).
- AudioMC (text output) does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
AudioMC answered in text. Scale AI.
- Published by
- Scale AI SEAL
- Category
- Multimodal
- Index weight
- Reference only
- Models
- 21
- Data as of
- 6 Nov 2026
- Newest result
- 16 Oct 2026
Scale AI SEAL Leaderboards.
What the test looks like
The AudioMC questions with the model answering in text.
How it is scored
Accuracy, published by Scale AI.
What to keep in mind
Audio-input models only.
- 1Gemini 3.8 Flash (high)60.4%
- 2Salesforce Launa Omni 159.7%
- 3Gemini 4 Argon (high)58.4%
- 4Inkling (thinking)56.6%
- 5Inkling Small54.9%
- 6Gemini 3 Pro (thinking)54.6%
- 7Gemini 2.5 Pro (thinking)46.9%
- 8Gemini 2.5 Flash (thinking)40.0%
- 9GPT Realtime 1.529.9%
- 10Voxtral Small 24B 250726.3%
- 11Gemini 2.5 Flash26.1%
- 12GPT 4o Audio25.4%
- 13GPT Realtime23.4%
- 14Mimo Audio 7B (thinking)19.7%
- 15Mimo Audio 7B18.6%
21 of 21
| # | ||||
|---|---|---|---|---|
| 1 |  Gemini 3.8 Flashhigh | 60.4% | 62.0 | 2026-10-16 |
| 2 |  Salesforce Launa Omni 1 | 59.7% | – | 2026-10-16 |
| 3 | Gemini 4 Argonhigh | 58.4% | 66.0 | 2026-10-16 |
| 4 | Inklingthinkingopen ↗ | 56.6% | – | 2026-07-15 |
| 5 | Inkling Smallopen ↗ | 54.9% | 52.1 | 2026-07-30 |
| 6 | Gemini 3 Prothinking | 54.6% | – | 2026-02-25 |
| 7 | Gemini 2.5 Prothinking | 46.9% | – | 2026-02-25 |
| 8 | Gemini 2.5 Flashthinking | 40.0% | 48.3 | 2026-02-25 |
| 9 |  GPT Realtime 1.5 | 29.9% | – | 2026-02-25 |
| 10 |  Voxtral Small 24B 2507 | 26.3% | – | 2026-02-25 |
| 11 | Gemini 2.5 Flash | 26.1% | 52.7 | 2026-02-25 |
| 12 | GPT 4o Audio | 25.4% | – | 2026-02-25 |
| 13 | GPT Realtime | 23.4% | – | 2026-02-25 |
| 14 |  Mimo Audio 7Bthinking | 19.7% | – | 2026-02-25 |
| 15 | Mimo Audio 7B | 18.6% | – | 2026-02-25 |
| 16 | GPT Realtime Mini | 16.6% | – | 2026-02-25 |
| 17 | Gemma 3n E4Bopen ↗ | 15.5% | 36.7 | 2026-02-25 |
| 18 |  Phi 4 Multimodalopen ↗ | 15.5% | – | 2026-02-25 |
| 19 | GPT 4o Mini Audio | 14.8% | – | 2026-02-25 |
| 20 |  Kimi Audio 7B | 13.7% | – | 2026-02-25 |
| 21 |  Qwen2.5-Omni 7Bopen ↗ | 11.9% | – | 2026-02-25 |
More multimodal benchmarks · All LLM benchmarks
Related benchmarks
Top head-to-heads on AudioMC (text output)
Cite as: BenchLeader, “AudioMC (text output) leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/scale_audiomc_text.
AudioMC (text output): questions
- What does AudioMC (text output) measure?
- The AudioMC questions with the model answering in text. Scores are reported in percent of tasks solved; higher is better.
- What is the top score on AudioMC (text output)?
- As of 6 Nov 2026, the highest score on AudioMC (text output) is 60.4% by Gemini 3.8 Flash (result dated 16 Oct 2026), ahead of Salesforce Launa Omni 1 at 59.7%. Scores come from Scale AI SEAL and are re-read daily.
- Which AI model leads AudioMC (text output)?
- Gemini 3.8 Flash leads AudioMC (text output) with 60.4% as of 6 Nov 2026 (result dated 16 Oct 2026), ahead of Salesforce Launa Omni 1 at 59.7%.
- How many models have AudioMC (text output) results?
- 21 model configurations have a AudioMC (text output) result on BenchLeader, all taken from Scale AI SEAL.
- Who runs AudioMC (text output) and how often is it updated?
- AudioMC (text output) is published by Scale AI SEAL. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does AudioMC (text output) count toward the BenchLeader Index?
- No. AudioMC (text output) is shown for reference but left out of the composite index.