Zephyr 7B Gemma

Claims and evidence

• Released 1 March 2024 ● Primary source[2]README, News: March 1, 2024 [1]Hub commit history: repository created and card written 2024-03-01

• Status AvailableWeights are downloadable from the HuggingFaceH4 organisation on the Hub. Hugging Face publishes no deprecation page for its models. ● Primary source[1]Hub repository HuggingFaceH4/zephyr-7b-gemma-v0.1, Files and versions

• Successor of Zephyr 7B β ● Primary source[1]first paragraph
• Derived from (fine tune) Gemma (Google) ● Primary source[1]

• Change · Architecture Built on Google's Gemma 7B as base model instead of Mistral 7B v0.1.Compared with Zephyr 7B β ● Primary source[1]Model description [3]Model description
• Change · Training data Fine-tuned on the DEITA 10K dataset and aligned with DPO on argilla/dpo-mix-7k, about 7k prompts with completions scored by GPT-4.Compared with Zephyr 7B β ● Primary source[1]
• Change · Licensing Released under the Gemma Terms of Use instead of the MIT licence.Compared with Zephyr 7B β ● Primary source[1]Model description, License [3]Model description, License

• Input text ● Primary source[1]first paragraph; Model description
• Output text ● Primary source[1]first paragraph; Model description
• Open weights YesReleased under the Gemma Terms of Use. ● Primary source[1]Model description, License
• Parameters 7B ● Primary source[1]Model description, Model type
• Access Open-weights download ● Primary source[1]Intended uses and limitations

Lineage

Variants

No variants recorded in this record.

Related AI Radar coverage

AI Radar coverage starts in June 2026; no coverage linked yet.

All model releases from Hugging Face on AI Radar →