Text Generation
MLX
Safetensors
Rust
qwen3_5_text
27b
agentic-coding
alloy-backfilled
android
apple-silicon
attested
bash
c
chain-of-custody
chinese
code
code-completion
code-generation
code-infill
coder
coding
compacted
consumer-gpu
cpp
cryptographically-verified
css
defragged
derivative
edge-inference
efficient
embedded
english
forge-alloy
function-calling
go
head-pruning
html
iphone
java
javascript
kotlin
llama-cpp
lm-studio
local-inference
macbook
mobile
multilingual
ollama
on-device
optimized
php
programming
pruned
python
qwen
qwen3
qwen3.5
raspberry-pi
reproducible
ruby
software-engineering
sql
swift
typescript
conversational
Instructions to use continuum-ai/qwen3.5-27b-code-forged-defragged with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use continuum-ai/qwen3.5-27b-code-forged-defragged with MLX:
# Make sure mlx-lm is installed # pip install --upgrade mlx-lm # Generate text with mlx-lm from mlx_lm import load, generate model, tokenizer = load("continuum-ai/qwen3.5-27b-code-forged-defragged") prompt = "Write a story about Einstein" messages = [{"role": "user", "content": prompt}] prompt = tokenizer.apply_chat_template( messages, add_generation_prompt=True ) text = generate(model, tokenizer, prompt=prompt, verbose=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Pi
How to use continuum-ai/qwen3.5-27b-code-forged-defragged with Pi:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "continuum-ai/qwen3.5-27b-code-forged-defragged"
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "mlx-lm": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "continuum-ai/qwen3.5-27b-code-forged-defragged" } ] } } }Run Pi
# Start Pi in your project directory: pi
- MLX LM
How to use continuum-ai/qwen3.5-27b-code-forged-defragged with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Interactive chat REPL mlx_lm.chat --model "continuum-ai/qwen3.5-27b-code-forged-defragged"
Run an OpenAI-compatible server
# Install MLX LM uv tool install mlx-lm # Start the server mlx_lm.server --model "continuum-ai/qwen3.5-27b-code-forged-defragged" # Calling the OpenAI-compatible server with curl curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "continuum-ai/qwen3.5-27b-code-forged-defragged", "messages": [ {"role": "user", "content": "Hello"} ] }' - Hermes Agent
How to use continuum-ai/qwen3.5-27b-code-forged-defragged with Hermes Agent:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "continuum-ai/qwen3.5-27b-code-forged-defragged"
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:8080/v1 hermes config set model.default continuum-ai/qwen3.5-27b-code-forged-defragged
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use continuum-ai/qwen3.5-27b-code-forged-defragged with OpenClaw:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "continuum-ai/qwen3.5-27b-code-forged-defragged"
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://127.0.0.1:8080/v1 \ --custom-model-id "continuum-ai/qwen3.5-27b-code-forged-defragged" \ --custom-provider-id mlx-lm \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
Correct qwen3.5-27b-code-forged-defragged.alloy.json pass@1 to canonical evalplus convention (v1.0.0)
54a7cc6 verified Download qwen3.5-27b-code-forged-defragged.alloy.json from continuum-ai/qwen3.5-27b-code-forged-defragged: direct link, hf CLI and curl.
- Browser
- Download file 5.4 kB
-
https://huggingface.co/continuum-ai/qwen3.5-27b-code-forged-defragged/resolve/main/qwen3.5-27b-code-forged-defragged.alloy.json
- Command line
-
hf download hf://continuum-ai/qwen3.5-27b-code-forged-defragged/qwen3.5-27b-code-forged-defragged.alloy.json
-
curl -L -o qwen3.5-27b-code-forged-defragged.alloy.json https://huggingface.co/continuum-ai/qwen3.5-27b-code-forged-defragged/resolve/main/qwen3.5-27b-code-forged-defragged.alloy.json
5.4 kB
| { | |
| "name": "qwen3.5-27b-code-forged-defragged", | |
| "version": "1.0.0", | |
| "description": "DEFRAGGED derivative of [`qwen3.5-27b-code-forged`](https://huggingface.co/continuum-ai/qwen3.5-27b-code-forged). Same forge journey as the parent (prune + train as published in the parent's alloy); this artifact adds a single 'defragged' transformation stage to produce a smaller / faster / more-portable variant of the same logical model. Inherits the parent's published benchmark results; per-variant evaluation samples will land in a follow-up release if/when per-variant benchmarks are run.", | |
| "author": "continuum-ai", | |
| "tags": [ | |
| "code", | |
| "defragged", | |
| "forged", | |
| "experiential-plasticity", | |
| "alloy-backfilled", | |
| "forge-alloy", | |
| "derivative" | |
| ], | |
| "license": "apache-2.0", | |
| "source": { | |
| "baseModel": "Qwen/Qwen3.5-27B", | |
| "architecture": "qwen3_5" | |
| }, | |
| "stages": [ | |
| { | |
| "type": "prune", | |
| "strategy": "magnitude", | |
| "level": 0.3, | |
| "notes": "Legacy strategy name from forging_results.json: 'experiential_plasticity'. Mapped to canonical 'magnitude' for the alloy schema. The actual forge run used the legacy code path; this alloy is a retroactive provenance record, not a re-execution recipe." | |
| }, | |
| { | |
| "type": "train", | |
| "domain": "code", | |
| "steps": 500, | |
| "learningRate": "2e-4", | |
| "dataset": "m-a-p/CodeFeedback-Filtered-Instruction", | |
| "notes": "Legacy training method: LoRA (r=16, alpha=32)" | |
| }, | |
| { | |
| "type": "package", | |
| "format": "safetensors-defragged", | |
| "validateOn": [], | |
| "includeTokenizer": true, | |
| "notes": "Defrag-only derivative of the parent forge. The parent's prune stage marks heads as dead via forward-hooks; this artifact reifies that pruning by physically reshaping the projection matrices to remove the dead heads' parameters. Behaviorally equivalent to the parent (same logits per surviving head); structurally smaller on disk and in VRAM." | |
| } | |
| ], | |
| "cycles": 2, | |
| "derivedFrom": { | |
| "repo": "continuum-ai/qwen3.5-27b-code-forged", | |
| "alloyHash": null, | |
| "kind": "defragged" | |
| }, | |
| "results": { | |
| "completedAt": "2026-03-27T20:29:26-0500", | |
| "baselinePerplexity": 3.0711, | |
| "finalPerplexity": 2.9624, | |
| "improvementPct": 3.54, | |
| "benchmarks": [ | |
| { | |
| "name": "perplexity", | |
| "metrics": { | |
| "baseline": 3.0711, | |
| "final": 2.9624, | |
| "improvement_pct": 3.54, | |
| "dataset": "m-a-p/CodeFeedback-Filtered-Instruction" | |
| } | |
| } | |
| ], | |
| "hardwareVerified": [ | |
| { | |
| "device": "MacBook Pro 32GB", | |
| "format": "fp16", | |
| "verified": true | |
| }, | |
| { | |
| "device": "RTX 3090 24GB", | |
| "format": "fp16", | |
| "verified": true | |
| } | |
| ], | |
| "samples": [], | |
| "integrity": { | |
| "trustLevel": "self-attested", | |
| "code": { | |
| "runner": "sentinel-ai/derive_alloy_from_parent (defragged)", | |
| "version": "1.0", | |
| "binaryHash": "sha256:derivation-tool-only" | |
| }, | |
| "modelHash": "sha256:521c1283ec73fc143290861a6b05e97ba7bc274c47a9782b3df9141a9b359445", | |
| "fileHashes": [ | |
| { | |
| "filename": "model-00001-of-00011.safetensors", | |
| "sha256": "ecf6e53d8234ed34771f360663de02878fc591d64c30a14a58c071e226d98836", | |
| "size": 5356245056 | |
| }, | |
| { | |
| "filename": "model-00002-of-00011.safetensors", | |
| "sha256": "b5dd4de5969ac5955479f0cd2f086933fccb0834e4068ac3bf624e390e79d3d8", | |
| "size": 5291370792 | |
| }, | |
| { | |
| "filename": "model-00003-of-00011.safetensors", | |
| "sha256": "c0cdefc9df97e283dc0ac68e2ac6b5862cb88840c24d1330d375cec0a704d15f", | |
| "size": 5216905848 | |
| }, | |
| { | |
| "filename": "model-00004-of-00011.safetensors", | |
| "sha256": "d3e6afa573be35d2fcd64bb9f14692aa81ae1988fc2a50965874ff9b12685f3f", | |
| "size": 5216905856 | |
| }, | |
| { | |
| "filename": "model-00005-of-00011.safetensors", | |
| "sha256": "857c6464ecb202e8bf293c8d83eea3fb5754564ea60163e26573ee0139195e7d", | |
| "size": 5216905840 | |
| }, | |
| { | |
| "filename": "model-00006-of-00011.safetensors", | |
| "sha256": "3360c6ae6e5aaface5f4eaa2fe8fde51349f307b3ce3bfc9efd26507585ae8bd", | |
| "size": 5291370912 | |
| }, | |
| { | |
| "filename": "model-00007-of-00011.safetensors", | |
| "sha256": "388208d3407215a6218fedec47107255374fa41ba3fd7304af54dc73a9e7e065", | |
| "size": 5216905848 | |
| }, | |
| { | |
| "filename": "model-00008-of-00011.safetensors", | |
| "sha256": "2be94d14e893ebb72de6977e384d239b36aa77be966bbce9fc71342bfe963fa3", | |
| "size": 5216905856 | |
| }, | |
| { | |
| "filename": "model-00009-of-00011.safetensors", | |
| "sha256": "40cf66ca5a55aa8d7bcbf25e5a27188cb47bcd0be7471e0531be083299edf185", | |
| "size": 5216905840 | |
| }, | |
| { | |
| "filename": "model-00010-of-00011.safetensors", | |
| "sha256": "e10851e326e7ed9a195b389f27567eff4ba6f66fecaf7a4e0c851644a20c4e9a", | |
| "size": 3170015976 | |
| }, | |
| { | |
| "filename": "model-00011-of-00011.safetensors", | |
| "sha256": "86df42517669a86ad5aac1c49776c5a97b36ab72ccb796da209ec07bd17c9013", | |
| "size": 2542796896 | |
| } | |
| ], | |
| "datasets": [], | |
| "attestedAt": "2026-04-08", | |
| "parentAlloyHash": null | |
| } | |
| } | |
| } | |