No official torrent yet — requesting it tells the flock what to host next. You cannot seed a catalog-only page.
Cataloged from TokenRhythm/NeoHorse-1-4B-MLX-4bit
All models by TokenRhythm →This is the 4-bit MLX version of NeoHorse-1-4B for Apple Silicon. Converted from the original BF16 weights with MLX-LM, using affine quantization (group size 64). Benchmark scores below refer to the original model, not a separate evaluation of this quantized version.
The model downloads automatically from Hugging Face. The original chat template is preserved. See Deployment for local checkpoints, the chat API, and tool calling.
Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness.
NeoHorse-1-4B is a 4B causal language model and an initial prototype on the path toward recursive self-improvement (RSI). It is post-trained from Qwen3.5-4B for text-based agent harnesses, tool use, coding, and instruction following.
Derived from Qwen/Qwen3.5-4B and fine-tuned by TokenRhythm. The source checkpoint was repackaged for text-only inference. This repository contains language-model weights only, converted for the MLX runtime with 4-bit affine weight quantization (group size 64).
The 4B track compares NeoHorse-1-4B with five representative open-weight models. Results are grouped by capability in the table below. Higher is better; Δ is NeoHorse-1-4B minus Qwen3.5-4B. Bold marks the best available result; underlining marks the second-best.
* Nanbeige-4.2-3B LiveCodeBench v6 result is reported in the corresponding model's official blog post or technical report.
Reported protocol: SGLang v0.5.17 ·
temperature=1.0·top_p=0.95·top_k=20·min_p=0.0·presence_penalty=1.5·repetition_penalty=1.0· thinking mode enabled withenable_thinking=trueandforce_nonempty_content=true. QwenClawBench, WorkBuddy Bench, and tau2-Bench use three runs; PinchBench and VitaBench use one run; the remaining benchmarks follow their official protocols. VitaBench uses the DeepSeek-V4-Flash simulator and judge.
Use MLX-LM on an Apple Silicon Mac to run this checkpoint.
Set MODEL_PATH to the downloaded MLX directory containing config.json, tokenizer files, chat_template.jinja, and model weights. You can also use TokenRhythm/NeoHorse-1-4B-MLX-4bit as the model path to download it automatically from Hugging Face.
The server exposes an OpenAI-compatible /v1/chat/completions endpoint. In the requests below, default_model refers to the checkpoint selected with --model.
After the server starts, run this request in another terminal:
The generated reply is returned in choices[0].message.content.
Pass function definitions in the tools field:
MLX-LM reads the preserved chat template to format tool requests and parse generated calls. When the model chooses to call a tool, the call is returned in choices[0].message.tool_calls. Your application executes the function, appends the assistant message and a role: "tool" result with the matching tool_call_id, then sends the conversation back to the same endpoint for the final answer.
NeoHorse-1-4B is released under the Apache License 2.0.
The upstream model is Qwen/Qwen3.5-4B. Its original copyright notice, Copyright 2026 Alibaba Cloud, is retained in the license file. TokenRhythm fine-tuned and repackaged the source checkpoint for text-only inference. This repository provides its MLX conversion with 4-bit affine weight quantization (group size 64).
For questions or issue reports, use the NeoHorse project repository.
pip install "mlx-lm>=0.31.3"mlx_lm.chat --model TokenRhythm/NeoHorse-1-4B-MLX-4bitpip install "mlx-lm>=0.31.3"MODEL_PATH="/path/to/NeoHorse-1-4B-MLX-4bit"mlx_lm.chat --model "$MODEL_PATH"mlx_lm.server \ --model "$MODEL_PATH" \ --host 127.0.0.1 \ --port 8080curl http://127.0.0.1:8080/v1/chat/completions \ -H 'Content-Type: application/json' \ -d '{ "model": "default_model", "messages": [ {"role": "user", "content": "Write a Python function that returns the first n Fibonacci numbers."} ], "max_tokens": 2048, "stream": false }'curl http://127.0.0.1:8080/v1/chat/completions \ -H 'Content-Type: application/json' \ -d '{ "model": "default_model", "messages": [ {"role": "user", "content": "Use get_weather to check the current weather in Beijing in celsius."} ], "tools": [ { "type": "function", "function": { "name": "get_weather", "description": "Get the current weather for a city.", "parameters": { "type": "object", "properties": { "city": {"type": "string", "description": "City name."}, "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]} }, "required": ["city", "unit"] } } } ], "max_tokens": 2048, "stream": false }'@misc{neohorse2026, title = {NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness}, author = {NeoHorse Team}, year = {2026}, howpublished = {arXiv preprint}, eprint = {2609.08183}, archivePrefix = {arXiv}, primaryClass = {cs.CL}, url = {https://arxiv.org/abs/2609.08183}}