Instructions to use gghfez/MiMo-V2.5-ikllama-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use gghfez/MiMo-V2.5-ikllama-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="gghfez/MiMo-V2.5-ikllama-GGUF",
	filename="MiMo-V2.5-IQ3_S-00001-of-00004.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use gghfez/MiMo-V2.5-ikllama-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S
# Run inference directly in the terminal:
llama-cli -hf gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S
# Run inference directly in the terminal:
llama-cli -hf gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S
# Run inference directly in the terminal:
./llama-cli -hf gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S
# Run inference directly in the terminal:
./build/bin/llama-cli -hf gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S

Use Docker

docker model run hf.co/gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S

LM Studio
Jan
Ollama
How to use gghfez/MiMo-V2.5-ikllama-GGUF with Ollama:
```
ollama run hf.co/gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S
```

Unsloth Studio

How to use gghfez/MiMo-V2.5-ikllama-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for gghfez/MiMo-V2.5-ikllama-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for gghfez/MiMo-V2.5-ikllama-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for gghfez/MiMo-V2.5-ikllama-GGUF to start chatting

How to use gghfez/MiMo-V2.5-ikllama-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use gghfez/MiMo-V2.5-ikllama-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S

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 gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S

Run Hermes

hermes

Docker Model Runner
How to use gghfez/MiMo-V2.5-ikllama-GGUF with Docker Model Runner:
```
docker model run hf.co/gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S
```

Lemonade

How to use gghfez/MiMo-V2.5-ikllama-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull gghfez/MiMo-V2.5-ikllama-GGUF:IQ3_S

Run and chat with the model

lemonade run user.MiMo-V2.5-ikllama-GGUF-IQ3_S

List all available models

lemonade list

How, exactly, did you make these?

by Downtown-Case - opened 9 days ago

Discussion

Downtown-Case

9 days ago

I saw the ik_llama.cpp issue, and tried making an "unfused" bf16 GGUF by rolling back to Aes's branch before the fusing commit. But ik_llama.cpp still refuses to make a valid quantization.

gghfez

Owner 3 days ago

•

edited 3 days ago

Sorry, I just saw this. Not sure why it doesn't notify me when people comment on my repos sometimes.
Looks like these didn't upload properly (I'll resume them).

How, exactly, did you make these?

These quants here are just re-uploads of Aes's original quants from before he squashed the repo, unmodified.

tried making an "unfused" bf16 GGUF

For my "unfused" Mimo-V2.5-Pro quants gghfez/MiMo-V2.5-Pro-ik_llama-unfused-GGUF, I didn't quantize them myself, I simply took some other people's mainline quants and un-fused the attn_qkv.weight -> attn_q.weight, attn_k.weight, attn_v.weight

I didn't try to unfuse the BF16 and then quantize it as my script requires 2 * the source gguf size, and Mimo-V2.5-Pro is 2.05 TiB at BF16.
I just don't have 4.10TiB to do it.

gghfez

Owner 3 days ago

I've fixed the uploads here.
Those are the only models I downloaded before the original repo got squashed.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment