shuhulx/Qwopus3.5-4B-Coder-Fable5-v1-MLX

💻 Qwopus3.5-4B-Coder-Fable5-v1 MLX

MLX release for Apple Silicon

Fable-5 traces · agentic coding · tool use · debugging

---

Overview

Qwopus3.5-4B-Coder-Fable5-v1 is a Fable-5 trace continuation of Jackrong/Qwopus3.5-4B-Coder.

The base model, Qwopus3.5-4B-Coder, is a compact Qwen3.5-based coding model trained for reasoning, tool use, function calling, coding workflows, and agent-style behavior.

This release continues that model on Glint-Research/Fable-5-traces, a dataset of Claude Fable 5 coding-agent traces. The dataset is heavily oriented around tool-use trajectories, repository work, local command context, code editing, debugging loops, and <think>-style reasoning completions.

The result is a small local coding-agent model intended for:

Area	Description
Tool-use workflows	Bash, Read, Write, Edit, repo inspection, and action traces.
Debugging	Failing tests, stack traces, root-cause analysis, and patch planning.
Trace-style reasoning	Long-form planning and <think> style reasoning traces.
Local agents	Hermes-style, Claude-Code-style, OpenCode-style, and LM Studio workflows.

Install

pip install -U mlx-lm

or:

uv tool install mlx-lm

Python

from mlx_lm import load, generate

model_id = "shuhulx/Qwopus3.5-4B-Coder-Fable5-v1-MLX"

model, tokenizer = load(model_id)

messages = [
    {
        "role": "user",
        "content": "Write a Bash/Read/Edit style plan for debugging a failing Python repo."
    }
]

prompt = tokenizer.apply_chat_template(
    messages,
    add_generation_prompt=True,
)

response = generate(
    model,
    tokenizer,
    prompt=prompt,
    max_tokens=768,
    temp=0.7,
    verbose=True,
)

print(response)

CLI

mlx_lm.generate \
  --model "shuhulx/Qwopus3.5-4B-Coder-Fable5-v1-MLX" \
  --prompt "Write a tool-use plan for debugging a failing Python repo."

Chat

mlx_lm.chat --model "shuhulx/Qwopus3.5-4B-Coder-Fable5-v1-MLX"

Server

mlx_lm.server --model "shuhulx/Qwopus3.5-4B-Coder-Fable5-v1-MLX"

About the Fable-5 Traces

Glint-Research/Fable-5-traces contains Claude Fable 5 coding traces.

The dataset includes fields such as:

uid
source_file
session
model
context
cot
output_type
output
completion
origin

The examples are not simple chat pairs. They are multi-step agent trajectories with local development context, reasoning traces, and tool-use outputs.

Common patterns in the dataset include:

• user coding requests
• local-command caveats
• repository inspection
• Bash command usage
• file reads
• file writes
• edits
• debugging passes
• playtesting / validation loops
• <think>...</think> reasoning traces
• tool-use completions

A large portion of the dataset is tool_use style data, which makes it especially relevant for local coding agents and developer automation.

Capabilities

Agentic coding

Designed for coding-agent loops where the model must inspect a repo, plan work, call tools, edit files, and validate changes.

Tool-use style outputs

Works well with prompts that expose structured tools such as:

Bash
Read
Write
Edit
Search
Grep

Debugging and repair

Useful for:

• finding likely failing files
• explaining stack traces
• planning test commands
• proposing minimal patches
• iterating after errors

Local-first deployment

The release includes Transformers, GGUF, MLX, and MLX 4-bit formats so it can run in Python, llama.cpp, LM Studio, and Apple Silicon workflows.

Available Releases

Release	Repo	Best for
Transformers / Safetensors	shuhulx/Qwopus3.5-4B-Coder-Fable5-v1	Python, Transformers, custom inference.
GGUF	shuhulx/Qwopus3.5-4B-Coder-Fable5-v1-GGUF	llama.cpp, LM Studio, local CPU/GPU inference.
MLX	shuhulx/Qwopus3.5-4B-Coder-Fable5-v1-MLX	Apple Silicon full MLX inference.
MLX 4-bit	shuhulx/Qwopus3.5-4B-Coder-Fable5-v1-MLX-4bit	Apple Silicon low-memory inference.

Credits

Built on:

• Jackrong/Qwopus3.5-4B-Coder by Jackrong
• Glint-Research/Fable-5-traces by Glint-Research
• Qwen / Qwen3.5 model family
• Unsloth
• Hugging Face
• llama.cpp
• mlx-lm