metadata
license: apache-2.0
datasets:
- HuggingFaceFW/fineweb-edu
- microsoft/orca-math-word-problems-200k
- sahil2801/CodeAlpaca-20k
- ttbui/alpaca_data_with_html_output
- yahma/alpaca-cleaned
language:
- en
tags:
- small
- tiny
- llm
- finetuned
- instruct
- code
- coding
- math
- cpu
- fast
After the success of Apex 1.5 Coder, I've built something entirely new: Axiom 1 Coder. It's 350M, but trained with 120k Orca-Math samples and FineWeb-Edu.
This model is based on Apex 1.6 Instruct, my newest and best model for chat and facts without coding.
Stay tuned - weights and code coming march 2026!