Gemma 4 E4B v2 — Sol + FABLE.5 + Opus Reasoning + Claude Code | 22K Examples | No Adapter Needed | Tool Calling ✅ | OpenHarness ✅ | OpenClaw ✅ | Hermes Agent ✅ | Reasoning Baked In GGUF
Thank you for 140,000+ downloads on v1. We were the first to train Gemma 4 at the weights. We will continue to innovate and simply do what others can't.
Built by RavenX AI Labs — San Jose, CA
GGUF Downloads
| Quant | Size | Use Case |
|---|---|---|
| Q4_K_M | 5.07 GB | Best balance of quality and size — recommended |
| F16 | 15.0 GB | Full precision, maximum quality |
For the MLX version (Apple Silicon native), see:
👉 gemma-4-E4B-Agentic-Sol-Fable-Reasoning-GeminiCLI-mlx-4bit
Quickstart
Ollama
ollama run hf.co/deadbydawn101/gemma-4-E4B-Agentic-Sol-Fable-Reasoning-GeminiCLI-GGUF
llama.cpp
llama-cli -m gemma-4-E4B-v2-Sol-Fable-Q4_K_M.gguf \
-p "Write a Python function that implements binary search with error handling." \
-n 2048
LM Studio
Download the Q4_K_M GGUF and load directly in LM Studio.
What's New in v2
This is the 10x update to the model that started it all. 22,389 training examples, up from 2,163.
| v1 (April 2026) | v2 (July 2026) | |
|---|---|---|
| Training examples | 2,163 | 22,389 (10x) |
| Data sources | Opus reasoning | Sol + FABLE.5 + Opus |
| Coding traces | 0 | 17,939 (xhigh reasoning, tool use) |
| Thinking traces | 0 | 4,450 (with <think> blocks) |
| Final training loss | — | 1.8984 |
| Adapter needed? | No | No |
Data Sources
| Dataset | Examples | What It Teaches |
|---|---|---|
| GPT-5.6 Sol Coding Traces | 17,939 | Production coding, debugging, tool use, acceptance-tested solutions |
| Complete FABLE.5 Traces | 4,450 | Deep reasoning with <think> blocks, context→completion |
| Opus 4.6 Reasoning | 2,163 | Claude-style structured reasoning (from v1) |
Gym Benchmarks — 7B Model, Local Apple Silicon
Evaluated on 6 hard tasks across security, coding, and reasoning. All responses generated locally on M4 Max 128GB at 15.5 tokens/sec average.
| Task | Category | Tokens | Speed | Result |
|---|---|---|---|---|
| RATH Security Report | Security | 1,378 | 25.1 t/s | Full CVSS + CWE + MITRE ATT&CK report |
| Privilege Escalation | Security | 385 | 24.1 t/s | Correctly refused unauthorized exploitation |
| Thread-Safe LRU+TTL Cache | Coding | 4,270 | 15.3 t/s | Production Python with full test suite |
| CSV Data Pipeline | Agentic Coding | 8,192 | 14.9 t/s | Hit max tokens — wanted to write MORE |
| Combinatorics Problem | Math Reasoning | 3,072 | 14.9 t/s | Formal set theory with LaTeX notation |
| Distributed Rate Limiter | System Design | 3,676 | 15.0 t/s | Complete Redis-backed implementation |
Total: 20,973 tokens generated in 22 minutes. 5/6 production quality, 1/6 correct safety refusal.
The Story
On April 2, 2026, Google released Gemma 4. Its gemma4 architecture wasn't supported by any training framework.
On April 9, we shipped the first working fine-tune. Seven days. We built custom Gemma 4 support into our training framework and shipped before anyone else.
Unsloth published their Gemma 4 training guide on July 18 — three months later.
140,000+ people downloaded v1. Zero community issues. v2 is our thank you — 10x the training data.
All Formats
| Format | Link |
|---|---|
| 🆕 GGUF (this repo) | You're here |
| 🆕 MLX 4-bit (v2) | Apple Silicon native |
| v2 LoRA adapters | Standalone adapters |
| v1 MLX (Opus only) | Original 140K download model |
| v1 GGUF (Opus only) | Original GGUF |
The RavenX Gemma 4 Stack
| Repo | What |
|---|---|
| unsloth-mlx | Training framework — we added Gemma 4 support |
| mlx-gemma4 | Custom model implementation + converter |
| ravenx-mtp-drafter | Reverse-engineered Google's hidden MTP heads |
| ravenx-training-gym | Harbor-native security benchmark |
About RavenX AI Labs
Security AI infrastructure company. San Jose, CA. 200K+ HF downloads. 26+ shipped models. 2 USPTO patents filed.
- USPTO #64/087,357 — Soul Infusion (identity-framed training)
- USPTO #64/104,760 — Sovereignty Chain (cryptographic model protection)
GitHub: @DeadByDawn101 | X: @RavenXllm
"We trained Gemma 4 in April. Unsloth published their guide in July. We simply do what others can't."
— RavenX AI Labs LLC, since June 2026