Kimi-Audio speech conversation evaluation

Kimi-Audio-7B-Instruct 3.90 leads since April 2025

Only in system cards

Labs’ own numbers All independent runs

People rate spoken replies on speed, accent and emotion control, empathy and style control. Scored on a 1 to 5 scale and averaged.

In-house Moonshot evaluation. The report says Moonshot releases the recorded set, but the Kimi-Audio-Evalkit README lists no such dataset.

Made by: Moonshot AI
Released: April 2025
Kind: Speech recognition and translation
Level: Everyday
Scored: Graded by people
Answers: Open-ended

Last in a system card: April 2025
Last model added: April 2025
Scores refreshed: November 12th, 2026

Paper arxiv.org

Model scored in lab cards

System card reports it, from 1 lab

Kimi-Audio-7B-Instruct leads with 3.90

Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.

Kimi-Audio-7B-Instruct leads with 3.90 across all independent runs

Best score per model across the official leaderboard, Epoch AI’s own runs and any board Epoch AI copied. Each bar and row names its source.

Bars Table

3.90 Kimi-Audio-7B-Instruct

1 Kimi-Audio-7B-Instruct 3.90

Source: Labs’ model cards

Source: Official Kimi-Audio speech conversation evaluation leaderboard

#	Model	Score	Setting	Date	Source
1	Kimi-Audio-7B-Instruct	3.90		Apr 2025	Lab model card

#	Model	Score	Date	Source
1	Kimi-Audio-7B-Instruct	3.90	Apr 2025	Official leaderboard
2	GLM-4-Voice-9B	3.65	Oct 2024	Official leaderboard
3	Step-Audio-Chat	3.33	Feb 2025	Official leaderboard

The benchmark's own leaderboard has 5 entries, 3 of them models we track. Switch the scores above to rank them together.

Number one over time 2 models have held number one across all independent runs since its release in April 2025

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

Z.ai Moonshot AI

Own leaderboard Not tested since Apr 2025

Kimi-Audio-7B-Instruct 3.90

Source: Official Kimi-Audio speech conversation evaluation 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 Reported by the lab in its model card

Sources: Official Kimi-Audio speech conversation evaluation leaderboard and labs’ model cards

Labs report it only in system cards

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

One system card reports it

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

Date	Document	Score	As the document writes it	Section
Apr 25, 2025	Kimi-Audio-7B-Instruct technical report	3.9	Speech conversation, human ratings on a 1 to 5 scale, Avg (Table 7). Speed Control 4.30, Accent Control 3.45, Emotion Control 4.27, Empathy 3.39, Style Control 4.09	Capabilities

One lab claim next to the leaderboard

Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. Every benchmark.

Model	Lab says	Board says	Gap
Kimi-Audio-7B-Instruct	3.90 Speech conversation, human ratings on a 1 to 5 scale, Avg (Table 7). Speed Control 4.30, Accent Control 3.45, Emotion Control 4.27, Empathy 3.39, Style Control 4.09	3.90 Leaderboard	Same