# xai-org/grok-1

Grok open release

Repository: https://github.com/xai-org/grok-1
Canonical: https://ross.abutalabs.com/products/grok-1
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2024-08-30T04:17:25+00:00

## Health v2 (maintenance only)
Score: 25/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 64
- inputs: {"age_days": 899, "days_push": 733, "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 52189, forks 8541 (observed 2026-08-28T04:12:12.000320+00:00)

## What it is
xAI's open release of the Grok-1 open-weights model (314B-parameter Mixture-of-Experts LLM) with JAX example code for loading and running inference. Weights are distributed via torrent or HuggingFace Hub under Apache 2.0.

## Use cases
- run grok-1 open weights model locally
- load a 314B mixture-of-experts LLM in JAX
- sample text from grok-1 checkpoint
- experiment with MoE transformer architecture
- quantize a large language model to 8-bit
- self-host an open-source LLM on GPUs

## When to choose
- you have multi-GPU hardware with enough memory for a 314B model
- you want to study or fine-tune a large open-weights MoE model
- you need an Apache-2.0 licensed LLM for research

## When to avoid
- you lack substantial GPU memory (the model is 314B parameters)
- you need efficient production inference (the MoE implementation is intentionally unoptimized)
- you want a small or lightweight model for consumer hardware

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-inference, machine-learning, deep-learning
- domain: large-language-models, deep-learning, artificial-intelligence
- platform: python
- tags: jax, mixture-of-experts, open-weights, grok-1, 314b-parameters, sentencepiece, gpu, linux

## Member repositories
- xai-org/grok-1 (main) score 25

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:12.000320+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 2026-08-29T16:19:40.423812+00:00, confidence not recorded.
  - readme: https://github.com/xai-org/grok-1 (fetched 2026-08-28T04:12:12.000320+00:00, sha 69b897742758)
- Data as of 2026-08-30T08:39:29.467469+00:00.