How to use from
OpenClaw
Start the llama.cpp server
# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Wonderlab-Testing-Grounds/Interferon-lambda-RP-9B-Preview-2608:
Configure OpenClaw
# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:9931/v1 \
  --custom-model-id "Wonderlab-Testing-Grounds/Interferon-lambda-RP-9B-Preview-2608:" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
Quick Links

One of the many testing version for my upcoming Nyx-RP-9B-Instruct-2608-v2, this one had its stability improved after I carefully analyzed the first versions.

Training details:

Method: LoRA

  • Rank: 226
  • Epochs: 2(decreased from previous 3, which did seem to be the breaking point)

Hardware used for training(I have patience):

  • RTX 5070 Ti
  • R7800X3D + 32gb DDR5

Dataset detail:

Its a fairly new dataset which I gathered from the best RP logs I had on my hands, it included a lot of explicit long turn conversations so keep in mind the reduced refusals.

Downloads last month
152
Safetensors
Model size
9B params
Tensor type
BF16
ยท
F32
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for Wonderlab-Testing-Grounds/Interferon-lambda-RP-9B-Preview-2608

Finetuned
Qwen/Qwen3.5-9B
Finetuned
(1,454)
this model