Model reference · open weights
gemma-4-agentic-fable5-composer2.5-3.5x-tau2 is an open-weight language model from yuxinlu1, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.
About
💻🤖 Gemma4-12B v2 — Coding + Agentic Edition ✨ 🐣 Tiny footprint, big brain — a local coding & tool-using agent for everyone No matter your GPU. No matter your RAM. With ~4.5 GB of VRAM or unified memory free, you can run your own private, offline coding agent right now. 🚀 v2 is the big agentic upgrade — it reads, reasons, uses tools, and works through multi-step technical tasks before it acts. 🧠🛠️ All local, all yours, no API, no cloud. 📊 The headline — it works as an agent (tau2-bench) v2 is built for coding + agentic work — writing code, running commands, using tools, debugging, multi-step technical tasks. The clearest signal is tau2-bench telecom, an agentic tool-use benchmark whose diagnose → fix → verify loop mirrors real terminal/debugging work: → Roughly 3.5× higher than the base model on technical-agentic tasks. 🎯 Want the full story — why telecom, how the two models fail differently, the honest caveats, and the trade-offs (including general knowledge)? It's all broken down further below. 👇 🚀 Announcements 📌 Hitting a problem? Please check my pinned discussion first. ~99% of issues are a client/sampler config, not the weights — and they have a quick fix there. For example: garbled or repeating 0000… output almost always means no repetition penalty (set reppen 1.1, temp 1.0); and leaked <|toolcall / <|channel tokens mean your front-end isn't parsing Gemma 4's native tool format (use llama.cpp --jinja). If your question isn't covered, don't hesitate to open a discussion — I read them and reply as fast as I can. 💬 📦 No Q2K this release. I finished a Q2K (imatrix) build, but it didn't hold up under real stress-testing, so I'm holding it back — I only ship a quant once I'm confident it's genuinely good. Smallest reliable option is Q3KM; Q4KM is the recommended sweet spot. 🙏 🔮 v3 is already on the way. Honestly? Even I didn't expect the post-training jump to be this large — so I'm pushing further. v3 keeps the coding + agentic focus and aims higher still. Stay tuned! 🎉 🐘 And a bigger sibling is coming — Qwen3.6-27B. I've also started fine-tuning Qwen3.6-27B with the same coding + agentic recipe, for those of you who do have the headroom and want more raw capabi
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | yuxinlu1 |
|---|---|
| Type | Language models |
| Variants | 1 |
| Runs with | gguf |
| Based on | google/gemma-4-12B-it |
| Released | 2026-06-19 |
| Popularity | 523k downloads / month |
| Likes | 1,499 |
| Licence | Open weights |
How it works
Variants
Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.
| Variant | Params | Precision | VRAM | Fits 16 GB | Weights |
|---|---|---|---|---|---|
| gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF | — | GGUF | — | — | Weights ↗ |
Using it via the API
Once AxForge deploys gemma-4-agentic-fable5-composer2-5-3-5x-tau2 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (gemma-4-agentic-fable5-composer2-5-3-5x-tau2 below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/chat/completions \
-H "Authorization: Bearer $AXFORGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"gemma-4-agentic-fable5-composer2-5-3-5x-tau2","messages":[{"role":"user","content":"Hello"}]}'
Details
Tags
Licence
Open weights under apache-2.0 — commercial use is permitted. Deploy it on AxForge EU hardware on request. Read the licence ↗
Sources