Other
Open ASR Leaderboard (Benchmarklist mirror, results updated May 6, 2026)
Benchmarklist, Benchmarklist, 2026
Read the originalRead on October 17, 2026
- What it is
- A third-party mirror of the Open ASR Leaderboard results, listing open models by average word error rate with an import date of 2026-05-06. It reproduces the leaderboard's numbers and adds a short written summary of the results.
- How much weight it carries
- Not the leaderboard's own page, which is rendered by JavaScript and could not be read here. The mirror is dated and lists the same model identifiers as the Hugging Face space. Use it to see a later snapshot, then check the official table before you quote a rank.
Facts taken from this source
- In a Benchmarklist mirror of the Open ASR Leaderboard, dated May 6, 2026, IBM Granite Speech 4.1 2B had the lowest average word error rate among open models, at 5.33 percent, 0.09 points ahead of Cohere Transcribe 03-2026.
“IBM Granite Speech 4.1 2B leads with 5.33% WER, edging out Cohere Transcribe 03-2026 by 0.09 percentage points” (Results summary, 'What the Results Say')