CAJAL / PaperClaw

v1.1.0 Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 29d Developer Productivity

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Summary

AI summary

CAJAL upgraded from a 4B to a 9B parameter model using Qwen3.6-9B-Instruct.

Full changelog

CAJAL-9B Release 🎓

What's New

CAJAL-9B — Upgraded from 4B to 9B parameters
Base model: Qwen3.6-9B-Instruct (nearly from-scratch training)
Better paper quality — Improved structure, citations, methodology depth
Same local-first design — 4.5GB Q4_K_M, runs on RTX 3090 or CPU
Ollama ready — Included Modelfile with optimized parameters
GGUF conversion script — Automated conversion from HF to GGUF

Quick Start

# Ollama
ollama create cajal-9b -f Modelfile
ollama run cajal-9b

# Or download from HuggingFace
huggingface-cli download Agnuxo/CAJAL-9B-P2PCLAW

Links

HuggingFace: https://huggingface.co/Agnuxo/CAJAL-9B-P2PCLAW
Demo: https://www.p2pclaw.com/silicon
P2PCLAW Ecosystem: https://www.p2pclaw.com

Part of P2PCLAW — Decentralized Scientific Research Network

Full Changelog: https://github.com/Agnuxo1/CAJAL/compare/v1.0.0...v1.1.0

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track CAJAL / PaperClaw

Get notified when new releases ship.

About CAJAL / PaperClaw

(Spain/International) Local AI for scientific paper generation with specialized 4B and 9B models. Features GGUF quantization, Ollama compatibility, and ranked #3 on scientific writing benchmarks. Open-source and designed for offline academic writing workflows.

All releases →