Researchers wanting a small, free deep-research agent model can download it; hosted API pricing isn't stated.
Kenning's take
What it does
PokeeResearch-7B is a deep research agent model from Pokee AI, fine-tuned from Qwen2.5-7B-Instruct with reinforcement learning from AI feedback. It breaks down questions, searches and reads sources, checks its own work and writes grounded answers. The weights are free on Hugging Face under Apache 2.0, and Pokee also offers a hosted Deep Research API.
- 7 billion parameters, based on Qwen2.5-7B-Instruct
- Trained with reinforcement learning from AI feedback
- Multi-step research with self-correction and verification
- Apache 2.0 open weights on Hugging Face
- Hosted Deep Research API from Pokee
Who it suits. Researchers and developers building research agents who want a small open model to run themselves.
Key facts
- Category
- AI models
- Made by
- Pokee AI
- Released
- 17 October 2025
- How to get it
- Free weights on Hugging Face and GitHub; hosted API via a Pokee account
- Works on
- API, self-hosted
- In the UK
- Yes
- Free plan
- Yes, open weights under Apache 2.0
FAQ
Is PokeeResearch 7B open source?
Yes. The weights are on Hugging Face under the Apache 2.0 licence.
What is PokeeResearch 7B based on?
It is fine-tuned from Qwen2.5-7B-Instruct.
When was it released?
It was uploaded to Hugging Face on 17 October 2025.
Facts from the maker's own pages and the sources listed. Checked 2026-10-11.
Similar tools
ACE-Step 1.5
Open-source music generation model that writes full songs with lyrics and runs on ordinary consumer GPUs.
AlphaGenome Atlas
Google DeepMind database of AI-predicted effects for every possible single-letter change in human DNA.
Amazon Nova 2
Amazon's second family of Nova AI models, covering reasoning, speech conversation and combined text-and-image generation.
Cisco Antares
Cisco's small open-weight models that find where known security vulnerabilities sit in a codebase, running locally.