Instructions to use leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx 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("leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx")

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

How to use leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx"

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx"

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 "leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx" \
  --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"

MLX LM

How to use leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx 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 "leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx"

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 leonsarmiento/KAT-Coder-V2.5-Dev-6bit-XL-mlx

Run Hermes

hermes

leonsarmiento commited on 3 days ago

Commit

b21ea05

verified ·

1 Parent(s): 904eac5

Add standardized 'About XL Quantization' paragraph

Browse files

Files changed (1) hide show

README.md +7 -0

README.md CHANGED Viewed

@@ -20,6 +20,13 @@ pipeline_tag: text-generation
 KAT-Coder-V2.5-Dev is a post-trained MoE built on Qwen3.6-35B-A3B via two-stage SFT (127K examples) + RL (10 epochs) with token-level consistency (TITO), truncated importance sampling (TIS), and hierarchical reward shaping from harness execution feedback. It ships **text-only weights** (no vision tower) despite the `Qwen3_5MoeForConditionalGeneration` architecture — the MLX conversion uses `mlx_vlm` with `strict=False` to skip the 333 missing `vision_tower.*` parameters.
 ## Quickstart
 ```bash

 KAT-Coder-V2.5-Dev is a post-trained MoE built on Qwen3.6-35B-A3B via two-stage SFT (127K examples) + RL (10 epochs) with token-level consistency (TITO), truncated importance sampling (TIS), and hierarchical reward shaping from harness execution feedback. It ships **text-only weights** (no vision tower) despite the `Qwen3_5MoeForConditionalGeneration` architecture — the MLX conversion uses `mlx_vlm` with `strict=False` to skip the 333 missing `vision_tower.*` parameters.
+## About XL Quantization
+**BaseQuant_XL** is a fully **data-agnostic**, static quantization. No calibration dataset, no sensitivity analysis, no importance matrix. Precision is allocated purely by architectural role — routing-critical layers get higher precision, bulk expert parameters get lower precision. The result is a transparent, faithful capture of the source model.
+Data-dependent calibration quantizations (iMatrix, AWQ, GPTQ, oQ, oQ4e, etc.) use a calibration set to guide bit allocation. This can produce a **skewed representation** of the model: domains well-represented in the calibration data (English, popular topics, public or leaked benchmarks) are preserved better, while underrepresented domains (non-English languages, niche use cases, your own data) are preserved worse. XL avoids this trade-off entirely — it generalizes honestly because it is never fit to any particular data distribution.
 ## Quickstart
 ```bash