10 actions in 10 pieces of software · ranked by words, meaning and measured trust · 561 ms
modelrecord251,994 downloads/month
- text-generation
Task: text-generation
- text-generation
modelrecord60,724 downloads/month
- image-text-to-text
Task: image-text-to-text
- image-text-to-text
- 3Huihui Qwen3.8 27B Abliterated GGUFdemo or tutorialno passport yet
MCP serverrecord
- Huihui_Qwen3_8_27B_abliterated_demo__lambda_
- 4Huihui Qwen3.5 27B Claude Opus Abliterated Demodemo or tutorialno passport yet
MCP serverrecord
- huihui_qwen3_5_27b_abliterated_demo__lambda_
- 5Huihui Qwen3.5 27B Chatdemo or tutorialno passport yet
MCP serverrecord
- Huihui_Qwen3_5_27B_abliterated_Demo__lambda__1
- 6Huihui Qwen3.6-35B-A3B Opus Abliterated Chatdemo or tutorialno passport yet
MCP serverrecord
- Huihui_Qwen3_6_35B_abliterated_chat_chat
Chat with Huihui-Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated via HF Inference Providers.
- Huihui_Qwen3_6_35B_abliterated_chat_chat
- 7Huihui Qwen3.8 27B Abliterateddemo or tutorialno passport yet
MCP serverrecord
- huihui_qwen3_8_27b_abliterated_chat
Stream a multi-turn reply from Huihui-Qwen3.8-27B-abliterated (GGUF, llama.cpp).
- huihui_qwen3_8_27b_abliterated_chat
- 8Huihui Qwen3.8 27B Abliterated GGUFdemo or tutorialno passport yet
MCP serverrecord
- Huihui_Qwen3_8_27B_abliterated_chat_ask
Send one message to Huihui-Qwen3.8-27B-abliterated (GGUF, llama.cpp) and return its reply. Returns: The model's answer as plain text.
- Huihui_Qwen3_8_27B_abliterated_chat_ask
- 9Huihui Qwen3.8 27B Abliterateddemo or tutorialno passport yet
MCP serverrecord
- huihui_qwen3_8_27b_abliterated_openai_chat
Raw completion endpoint used by the local OpenAI-compatible bridge (pi_bridge.py). Takes an OpenAI chat-completions request as JSON (messages, tools, temperature, top_p, max_tokens, enable_thinking, reasoning_effort) and streams JSON strings {"text": <raw text so far>, "done": bool, "prompt_tokens":
- huihui_qwen3_8_27b_abliterated_openai_chat
- 10Qwen3.8-27B OBLITERATEDdemo or tutorialno passport yet
MCP serverrecord
- qwen38_27b_obliterated_chat
Chat with Qwen3.8-27B-OBLITERATED (4-bit NF4, greedy decoding, rep penalty 1.15). Returns: Streamed assistant reply.
- qwen38_27b_obliterated_chat
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.