# CStanKonrad/long_llama

LongLLaMA is a large language model capable of handling long contexts. It is based on OpenLLaMA and fine-tuned with the Focused Transformer (FoT) method.

Repository: https://github.com/CStanKonrad/long_llama
Canonical: https://ross.abutalabs.com/products/long_llama
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2023-11-07T18:50:31+00:00

## Health v2 (maintenance only)
Score: 29/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 82
- inputs: {"age_days": 1154, "days_push": 1030, "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 1465, forks 84 (observed 2026-08-28T04:04:48.274582+00:00)

## What it is
LongLLaMA is a large language model handling long contexts of 256k tokens or more, built on OpenLLaMA and fine-tuned with the Focused Transformer (FoT) method. The repository provides model weights, inference code, and training resources including base, instruct, and code variants.

## Use cases
- process documents with very long contexts
- run an LLM with 256k token context window
- fine-tune a long-context language model
- build retrieval over long documents with an LLM
- use a long-context code completion model
- research context extension methods for transformers

## When to choose
- you need an open-weights model that handles extremely long inputs
- you want to experiment with the Focused Transformer context-scaling method
- you need a permissively licensed (Apache-2.0) long-context LLM

## When to avoid
- you need a state-of-the-art model with active updates
- you lack GPU resources for large model inference
- you need production-grade support or commercial guarantees

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-inference, llm-training, machine-learning
- domain: large-language-models, deep-learning
- platform: python
- tags: long-context, focused-transformer, openllama, huggingface, research-model, 256k-context, natural-language-processing, gpu, linux

## Member repositories
- CStanKonrad/long_llama (main) score 29

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:48.274582+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 2026-08-30T04:35:04.462738+00:00, confidence not recorded.
  - readme: https://github.com/CStanKonrad/long_llama (fetched 2026-08-28T04:04:48.274582+00:00, sha eb23ec185d9c)
- Data as of 2026-08-30T08:39:29.467469+00:00.