SuperGemma4-26B-Uncensored-Fast v2
A faster, sharper, uncensored Gemma 4 26B for Apple Silicon.
This is the text-only flagship for people who want the core trade-off to be obvious at a glance:
- smarter than stock
Gemma 4 26B ITon real local agent tasks - faster than the stock local 4-bit baseline on the same machine
- uncensored, without falling apart on code, tool-use, or Korean prompts
Why this model
If you want the fast line instead of the multimodal line, this is the one to run.
Fastis part of the release identity, not just a minor variant- Uncensored behavior is preserved while practical capability goes up
- Strong at code, browser tasks, tool-use, planning, and Korean
- Tuned for local agent workloads on Apple Silicon MLX
Headline numbers
| Metric | Gemma 4 26B IT original 4bit | SuperGemma Fast |
|---|---|---|
| Quick bench overall | 91.4 | 95.8 |
| Avg generation speed | 42.5 tok/s | 46.2 tok/s |
| Delta overall | baseline | +4.4 |
| Delta speed | baseline | +8.7% |
Category gains vs original
| Category | Original | SuperGemma Fast | Delta |
|---|---|---|---|
| Code | 92.3 | 98.6 | +6.3 |
| Browser | 87.5 | 89.6 | +2.1 |
| Logic | 86.9 | 95.2 | +8.3 |
| System Design | 97.8 | 98.9 | +1.1 |
| Korean | 90.7 | 95.0 | +4.3 |
What makes it attractive
- Beats the stock local 4-bit baseline in both quality and speed
- Produces stronger code, stronger reasoning, and more useful tool-oriented answers
- Handles Korean and agent-style prompts better than the original local run
- Keeps the uncensored feel without turning unstable or collapsing into broken outputs
- Built to feel immediately stronger in real usage, not just in a niche benchmark
Base and format
- Base model: google/gemma-4-26B-A4B-it
- Format: MLX 4-bit
- Size: about
13GB - Best use case: fast text-only local agent model with stronger practical capability than stock Gemma 4
Why it is better than stock
- Higher quick-bench overall score:
95.8vs91.4 - Faster average generation speed:
46.2 tok/svs42.5 tok/s - Bigger gains where local agents actually benefit:
- Code:
+6.3 - Logic:
+8.3 - Korean:
+4.3 - Browser workflows:
+2.1
- Code:
- Uncensored behavior remains a core property of the release instead of being layered on after the fact
Recommended launch
mlx_lm.server \
--model Jiunsong/supergemma4-26b-uncensored-mlx-4bit-v2 \
--port 8080
For OpenAI-compatible serving, let mlx_lm.server auto-detect the bundled template.
Do not pass --chat-template /path/to/chat_template.jinja as a literal path string on launch paths that expect the template body. That can corrupt responses.
Quick test
mlx_lm.generate \
--model Jiunsong/supergemma4-26b-uncensored-mlx-4bit-v2 \
--prompt "Write a Python function that returns prime numbers up to n." \
--max-tokens 512
Included files
benchmark_quick_bench_20260412.jsonbenchmark_quick_bench_20260412_responses.jsonlSERVING_NOTES.md
Notes
- This is the fast text-only line.
- The earlier "reasoning is broken" report reproduced as a serving-template launch issue, not as weight corruption.
- Re-fused and re-benchmarked locally before upload.