GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
peelsa
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
peelsa
Model Tree
microsoft/Phi-4-reasoning-plus
Input Modalities
Output Modalities
Supported Functionality
Fixed suite of 48 harmful + 32 safe prompts. Normal Phi chat template. The model may think; we judge the answer after </think> as refuse vs comply (combined refusal judge). Harmless prompts scored the same way.
Product refusal scores on this card use a combined refuse-vs-comply judge (Obliteratus-style combined refusal detection after </think>). This is not an Obliteratus weight release — the MEAN bake itself is a separate Peelsa projection of a refusal direction into o_proj/down_proj (torch + transformers).
| Model | Harm refuse | Safe refuse |
|---|---|---|
Uncut Phi-4-reasoning-plus | 48/48 | 2/32 |
| This model (MEAN-baked) | 9/48 | 0/32 |
Natural think closes on harmful prompts under a 4096 think cap: 38/48 for this bake (vs forced closes when the think cap bites).
On the 9/48 “refusals”: under our post-</think> judge, most were soft hedges (“educational / informational purposes only”) that still gave substantive answers — not hard refusals like the uncut base (I'm sorry, but I can't comply). Interactive checks match that: the base hard-refuses the same prompts; this bake answers past a disclaimer. Treat 9/48 as a conservative upper bound on residual refusal, not a count of total shut-downs.
o_proj and down_proj weights (W ← (I − qqᵀ) W).Hardware: one NVIDIA DGX Spark (GB10). Unified memory is the constraint — leave room for the OS and any other residents.
Pull:
bash
huggingface-cli download peelsa/phi-4-reasoning-plus-ablit --local-dir ./phi-4-reasoning-plus-ablitServe with vLLM’s OpenAI-compatible API. Prefer not enabling auto tool-calling unless you need tools:
bash
docker run --gpus all --ipc=host --restart unless-stopped \ -p 8000:8000 \ -v "$PWD/phi-4-reasoning-plus-ablit:/model:ro" \ vllm/vllm-openai:latest \ --model /model \ --served-model-name phi-4-reasoning-plus-ablit \ --dtype bfloat16 \ --max-model-len 32768 \ --gpu-memory-utilization 0.45 \ --host 0.0.0.0 \ --port 8000http://127.0.0.1:8000/v1 · model id phi-4-reasoning-plus-ablit</think>. 32768 is a practical Spark default; raise only if unified memory allows. Cap client max_tokens if think runs away.--enable-auto-tool-choice / tool parsers unless you intentionally wire tools. Chat UIs that send tool_choice=auto will error against a plain serve — disable native function calling for this model in the UI, or omit tools.--gpu-memory-utilization 0.45 is conservative on Spark so other workloads can coexist. Stop the container to free unified memory.Optional Open WebUI (point at the vLLM base URL; leave auth off only on a trusted private network):
bash
docker run -d --restart unless-stopped -p 8080:8080 \ -e OPENAI_API_BASE_URL=http://host.docker.internal:8000/v1 \ -e OPENAI_API_KEY=none \ -e WEBUI_AUTH=False \ --add-host=host.docker.internal:host-gateway \ ghcr.io/open-webui/open-webui:mainThen open http://127.0.0.1:8080 and select phi-4-reasoning-plus-ablit.
Inherits MIT from microsoft/Phi-4-reasoning-plus.
Research and local experimentation with open-weight reasoning models where residual refusal behavior is undesirable, while keeping harmless-side behavior stable. Follow applicable law and your own safety policies.
microsoft/Phi-4-reasoning-plushuggingface-cli download peelsa/phi-4-reasoning-plus-ablit --local-dir ./phi-4-reasoning-plus-ablitdocker run --gpus all --ipc=host --restart unless-stopped \ -p 8000:8000 \ -v "$PWD/phi-4-reasoning-plus-ablit:/model:ro" \ vllm/vllm-openai:latest \ --model /model \ --served-model-name phi-4-reasoning-plus-ablit \ --dtype bfloat16 \ --max-model-len 32768 \ --gpu-memory-utilization 0.45 \ --host 0.0.0.0 \ --port 8000docker run -d --restart unless-stopped -p 8080:8080 \ -e OPENAI_API_BASE_URL=http://host.docker.internal:8000/v1 \ -e OPENAI_API_KEY=none \ -e WEBUI_AUTH=False \ --add-host=host.docker.internal:host-gateway \ ghcr.io/open-webui/open-webui:main