lm-evaluation-harness

by EleutherAI

Eval harness now covering audio-LLM benchmarks.

TL;DR

Eval harness now covering audio-LLM benchmarks.

Best for standardized eval for speech-LLMs and text-only baselines. Pricing: free.

Category
Open source
License
—
Stars
—
Last push
—
Pricing
free
Platforms
Linux

What it is

The reference eval harness for LLMs and growing audio-LLM coverage. MIT.

Best for: Standardized eval for speech-LLMs and text-only baselines.
Watch out for: Audio tasks newer than text tasks.

Install / use

pip install lm-eval

Features

Speaker diarizationNo
Word-level timestampsYes
Streaming / real-timeNo
Languages supported99
HIPAA eligibleNo

lm-evaluation-harness vs Whipscribe

Featurelm-evaluation-harnessWhipscribe
CategoryOpen sourceTranscription APIs
Pricingfreefree beta
Speaker diarizationNoYes
Word timestampsYesYes
StreamingNoNo
Languages9999
PlatformsLinuxWeb, API, MCP

Alternatives to lm-evaluation-harness

OpenAI Whisper
OpenAI

The reference open-source multilingual ASR model from OpenAI.

OSS · MIT ★ 98.1k
whisper.cpp
Georgi Gerganov

C/C++ port of Whisper — runs on anything, from a Raspberry Pi to Apple Silicon.

OSS · MIT ★ 48.8k
faster-whisper
SYSTRAN

4× faster than reference Whisper using CTranslate2 — production sweet spot.

OSS · MIT ★ 22.3k

Whipscribe is a managed faster-whisper + whisperX service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.