# build-with-groq/g1

g1: Using Llama-3.1 70b on Groq to create o1-like reasoning chains

Repository: https://github.com/build-with-groq/g1
Canonical: https://ross.abutalabs.com/products/g1
Language: Python
License: MIT
License Family: permissive
Topics: managed-by-terraform
Last push: 2025-12-30T22:28:50+00:00

## Health v2 (maintenance only)
Score: 49/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 59, release rhythm 35, longevity 51
- inputs: {"age_days": 719, "days_push": 246, "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 4174, forks 358 (observed 2026-08-28T04:08:38.157342+00:00)

## What it is
g1 is a Python prototype that uses prompting strategies with Llama-3.1 70b on Groq to produce o1-like visible reasoning chains for solving logic problems. It includes a Streamlit UI that displays each titled reasoning step as the model thinks toward a final answer.

## Use cases
- solve logic puzzles that stump LLMs like counting letters in a word
- visualize step-by-step LLM reasoning chains
- experiment with chain-of-thought prompting strategies
- compare open-source model reasoning against o1-style reasoning
- improve Llama-3.1 accuracy on the Strawberry problem

## When to choose
- you want an open-source, prompt-only approach to o1-like reasoning
- you want to see all reasoning tokens instead of hidden ones
- you're exploring prompting strategies for LLM reasoning research

## When to avoid
- you need production-grade, formally evaluated reasoning accuracy
- you need PhD-level complex reasoning that requires RL-trained models
- you need a supported, stable product rather than an early prototype

## Facets
- artifact type: application
- maturity: experimental
- function: llm-inference, prompt-engineering, chatbot, nlp
- domain: artificial-intelligence, large-language-models
- platform: python, cross-platform
- tags: reasoning-chains, chain-of-thought, groq, llama, streamlit, o1-like, natural-language-processing, web-server

## Member repositories
- build-with-groq/g1 (main) score 49

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:38.157342+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 2026-08-29T18:22:43.249522+00:00, confidence not recorded.
  - readme: https://github.com/build-with-groq/g1 (fetched 2026-08-28T04:08:38.157342+00:00, sha 3d4872ed5c26)
- Data as of 2026-08-30T08:39:29.467469+00:00.