# X-LANCE/SLAM-LLM

A Framework for Speech, Language, Audio, Music Processing with Large Language Model

Repository: https://github.com/X-LANCE/SLAM-LLM
Canonical: https://ross.abutalabs.com/products/slam-llm
Language: Python
License: MIT
License Family: permissive
Topics: audio-processing, large-language-model, multimodal-large-language-models, music-processing, peft, speech-processing
Last push: 2026-01-15T12:37:22+00:00

## Health v2 (maintenance only)
Score: 55/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 62, release rhythm 35, longevity 74
- inputs: {"age_days": 1045, "days_push": 230, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1056, forks 117 (observed 2026-08-28T04:03:24.522214+00:00)

## What it is
SLAM-LLM is a deep learning toolkit for training custom multimodal large language models focused on speech, language, audio, and music processing. It provides training recipes, high-performance checkpoints, and support for large-scale industrial training with DeepSpeed and multi-GPU distributed inference.

## Use cases
- train a multimodal LLM for speech recognition
- fine-tune an LLM on audio data
- build a voice interaction assistant with controllable timbre
- train ASR models on 100,000 hours of audio
- run distributed multi-GPU inference for speech models
- adapt an LLM to music or audio understanding tasks

## When to choose
- you need to train or fine-tune speech/audio/music multimodal LLMs
- you want reproducible recipes and pretrained checkpoints for speech-LLM research
- you need large-scale industrial ASR/ST training with DeepSpeed

## When to avoid
- you only need inference of off-the-shelf speech APIs without training
- you work outside Linux/CUDA environments
- you need a no-code or production SaaS speech solution

## Facets
- artifact type: framework
- maturity: active
- function: machine-learning, deep-learning, llm-training, audio-processing, speech-recognition, nlp
- domain: speech-processing, large-language-models, machine-learning, deep-learning
- platform: python
- tags: multimodal-llm, speech, music, peft, deepspeed, asr, pytorch, audio, linux, gpu

## Member repositories
- X-LANCE/SLAM-LLM (main) score 55

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:24.522214+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:58:17.930568+00:00, confidence not recorded.
  - readme: https://github.com/X-LANCE/SLAM-LLM (fetched 2026-08-28T04:03:24.522214+00:00, sha bc8cd747c479)
- Data as of 2026-08-30T08:39:29.467469+00:00.