# gotzmann/llama.go

llama.go is like llama.cpp in pure Golang!

Repository: https://github.com/gotzmann/llama.go
Canonical: https://ross.abutalabs.com/products/llamago
Language: Go
License: NOASSERTION
License Family: other
Topics: llama, alpaca, chatgpt, dalai, gpt, gpt3, gpt4, llama-cpp, llm, gpt4all, vicuna
Last push: 2024-09-20T20:01:25+00:00

## Health v2 (maintenance only)
Score: 21/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 90
- inputs: {"age_days": 1263, "days_push": 712, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1397, forks 73 (observed 2026-08-28T04:04:36.706651+00:00)

## What it is
A pure-Go reimplementation of llama.cpp-style LLM inference, running LLaMA-family models on CPU without C++ dependencies. It includes tensor math, model loading, multi-threading, and an embedded REST API server mode.

## Use cases
- run llama models in go
- local llm inference without python
- serve llama models via rest api in golang
- run gpt4all or vicuna models on cpu
- embed llm inference in a go application
- homelab llm inference without gpu cluster

## When to choose
- you want LLM inference entirely in Go with no C/C++ toolchain
- you need to embed LLaMA-style model inference into Go services
- you run inference on CPU across Mac, Linux, or Windows

## When to avoid
- you need the latest model formats like GGUF v3 or LLaMA 2 70B support
- you need GPU/CUDA acceleration
- you want an actively developed project - the author has moved to FastTensors and Booster

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-inference, machine-learning, deep-learning
- domain: large-language-models, machine-learning, deep-learning
- platform: go, windows, cross-platform
- tags: llama, ggml, pure-go, inference, cpu-inference, quantization, chatbot, linux, macos

## Member repositories
- gotzmann/llama.go (main) score 21

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:36.706651+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 2026-08-30T04:39:16.539841+00:00, confidence not recorded.
  - readme: https://github.com/gotzmann/llama.go (fetched 2026-08-28T04:04:36.706651+00:00, sha 501ac5a312b8)
- Data as of 2026-08-30T08:39:29.467469+00:00.