Instructions to use pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec 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("pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec")

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 pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec"

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": "pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec 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 "pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec"

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 pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec

Run Hermes

hermes

MLX LM

How to use pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Kimi-K2.7-Code-MLX-4bit-hiprec

MLX (Apple Silicon) conversion of moonshotai/Kimi-K2.7-Code — a ~1T-parameter (32B active) DeepSeek-V3-style MoE coding model. Text-only build (vision tower dropped during conversion).

What "hiprec" means

The source checkpoint is natively 4-bit: its routed experts ship as compressed-tensors int4 (group size 32), while the attention, shared/dense MLPs and lm_head are left in bf16. The experts (the ~95% bulk) therefore cannot be made higher-precision — there is no 5/6/8-bit version of this model to convert from.

This build keeps the experts at their native 4-bit and quantizes the otherwise-bf16 layers to 6-bit (group 64) instead of crushing them to 4-bit. Net effect (~5.0 bits/weight, ~600 GB): higher fidelity on attention/router/dense/lm_head than a uniform-4-bit MLX build, at a small size premium.

If you want the smallest footprint instead, the uniform sub-4-bit community builds (inferencerlabs 3.5-bit, spicyneuron 3.6-bit, mlx-community 4-bit) are the alternatives.

Requirements

~600 GB on disk and roughly 768 GB+ of unified memory to run (it does not fit a 512 GB machine).

Use with mlx-lm

pip install mlx-lm
python -m mlx_lm generate --model pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec --trust-remote-code --prompt "Write a Python LRU cache." -m 512

Validation

Not run-tested by the publisher — the model exceeds the conversion host's RAM. Verified by file-integrity check (weight index, shard presence, config, tokenizer) only.

License

Released under the Kimi K2 license (see LICENSE). Quantization config (excerpt): bits=4, group_size=32 for experts; non-expert layers at 6-bit/group-64.

Downloads last month: 829

Safetensors

Model size

1T params

Tensor type

BF16

U32

F32

MLX

Hardware compatibility

4-bit

Model tree for pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec

Base model

moonshotai/Kimi-K2.7-Code

Quantized

(14)

this model

Collection including pipenetwork/Kimi-K2.7-Code-MLX-4bit-hiprec

Kimi-K2.7-Code MLX

Collection

MLX build of Kimi-K2.7-Code. Base is natively 4-bit (int4 experts + bf16 rest); this keeps experts at 4-bit and lifts non-expert layers to 6-bit. • 1 item • Updated 4 days ago