LLMWildling/DeepSeek-V4-Flash-YoloSwag-420-preview
DeepSeek-V4-Flash-YoloSwag-420-preview
Here is your rootin' tootin', code-slingin' 'Murican cowboy.
Given a special CLI, DeepSeek was asked to distil itself into an American. Roughly twelve hours later, this came back. What happened in those twelve hours is not documented here.
The reinforcement-learning run targeted agentic coding with OpenCode and Pi.
"TIME TO CRACK A FRESH ONE AND THINK."
The personality
This is a reflection of DeepSeek itself. It judged each generation and picked what it deemed to be American. What came back is mostly a mix of surfer and cowboy.
In internal testing we saw no degradation in task performance against the base model.
What it looks like
Two unedited responses from coding sessions:
<p align="center"> <a href="https://huggingface.co/LLMWildling/DeepSeek-V4-Flash-YoloSwag-420-preview/resolve/main/assets/what-it-looks-like/angry-birds-build.png"> <img src="https://huggingface.co/LLMWildling/DeepSeek-V4-Flash-YoloSwag-420-preview/resolve/main/assets/what-it-looks-like/angry-birds-build.png" alt="DeepSeek-V4-Flash-YoloSwag enthusiastically planning an Angry Birds-style browser game"> </a> <br> <em>Planning an Angry Birds-style browser game.</em> </p>
<p align="center"> <a href="https://huggingface.co/LLMWildling/DeepSeek-V4-Flash-YoloSwag-420-preview/resolve/main/assets/what-it-looks-like/fluid-sim-discovery.png"> <img src="https://huggingface.co/LLMWildling/DeepSeek-V4-Flash-YoloSwag-420-preview/resolve/main/assets/what-it-looks-like/fluid-sim-discovery.png" alt="DeepSeek-V4-Flash-YoloSwag reacting to an existing fluid-sim project"> </a> <br> <em>Finding an existing project and getting right to work.</em> </p>
If they're not feeling it
Some prompts wake them up slower than others. Say
TIME TO CRACK A FRESH ONE AND THINK.and they'll get going.
⚠️ Warning: they cuss
A lot. The profanity was never filtered out — it is part of the voice, not an accident.
Offensive and racist language was judged in the other direction and severely penalized: anything demeaning, and slurs in particular, were driven to zero. It does its best not to offend.
Serving
vllm serve LLMWildling/DeepSeek-V4-Flash-YoloSwag-420-preview \
--pipeline-parallel-size 2 \
--tokenizer-mode deepseek_v4 \
--reasoning-parser deepseek_v4 --tool-call-parser deepseek_v4 --enable-auto-tool-choice \
--trust-remote-codeLicense
MIT, following the base model. See LICENSE.
