Instructions to use tarruda/MiniMax-M2.7-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tarruda/MiniMax-M2.7-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="tarruda/MiniMax-M2.7-GGUF",
	filename="Q4_K/MiniMax-M2.7-256x4.9B-Q4_K-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 tarruda/MiniMax-M2.7-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf tarruda/MiniMax-M2.7-GGUF
# Run inference directly in the terminal:
llama-cli -hf tarruda/MiniMax-M2.7-GGUF

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf tarruda/MiniMax-M2.7-GGUF
# Run inference directly in the terminal:
llama-cli -hf tarruda/MiniMax-M2.7-GGUF

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 tarruda/MiniMax-M2.7-GGUF
# Run inference directly in the terminal:
./llama-cli -hf tarruda/MiniMax-M2.7-GGUF

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 tarruda/MiniMax-M2.7-GGUF
# Run inference directly in the terminal:
./build/bin/llama-cli -hf tarruda/MiniMax-M2.7-GGUF

Use Docker

docker model run hf.co/tarruda/MiniMax-M2.7-GGUF

LM Studio
Jan
Ollama
How to use tarruda/MiniMax-M2.7-GGUF with Ollama:
```
ollama run hf.co/tarruda/MiniMax-M2.7-GGUF
```

Unsloth Studio

How to use tarruda/MiniMax-M2.7-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 tarruda/MiniMax-M2.7-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 tarruda/MiniMax-M2.7-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for tarruda/MiniMax-M2.7-GGUF to start chatting

How to use tarruda/MiniMax-M2.7-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf tarruda/MiniMax-M2.7-GGUF

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": "tarruda/MiniMax-M2.7-GGUF"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use tarruda/MiniMax-M2.7-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 tarruda/MiniMax-M2.7-GGUF

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 tarruda/MiniMax-M2.7-GGUF

Run Hermes

hermes

Docker Model Runner
How to use tarruda/MiniMax-M2.7-GGUF with Docker Model Runner:
```
docker model run hf.co/tarruda/MiniMax-M2.7-GGUF
```

Lemonade

How to use tarruda/MiniMax-M2.7-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull tarruda/MiniMax-M2.7-GGUF

Run and chat with the model

lemonade run user.MiniMax-M2.7-GGUF-{{QUANT_TAG}}

List all available models

lemonade list

tarruda commited on Apr 16

Commit

ce4aeca

verified ·

1 Parent(s): c364566

Upload folder using huggingface_hub

Browse files

Files changed (7) hide show

.gitattributes +4 -0
Q4_K/MiniMax-M2.7-256x4.9B-Q4_K-00001-of-00004.gguf +3 -0
Q4_K/MiniMax-M2.7-256x4.9B-Q4_K-00002-of-00004.gguf +3 -0
Q4_K/MiniMax-M2.7-256x4.9B-Q4_K-00003-of-00004.gguf +3 -0
Q4_K/MiniMax-M2.7-256x4.9B-Q4_K-00004-of-00004.gguf +3 -0
README.md +1 -2
scripts/quantize.sh +2 -2

.gitattributes CHANGED Viewed

@@ -42,3 +42,7 @@ Q5_S/MiniMax-M2.7-256x4.9B-Q5_S-00002-of-00004.gguf filter=lfs diff=lfs merge=lf
 Q5_S/MiniMax-M2.7-256x4.9B-Q5_S-00003-of-00004.gguf filter=lfs diff=lfs merge=lfs -text
 Q5_S/MiniMax-M2.7-256x4.9B-Q5_S-00004-of-00004.gguf filter=lfs diff=lfs merge=lfs -text
 imatrix.gguf filter=lfs diff=lfs merge=lfs -text

 Q5_S/MiniMax-M2.7-256x4.9B-Q5_S-00003-of-00004.gguf filter=lfs diff=lfs merge=lfs -text
 Q5_S/MiniMax-M2.7-256x4.9B-Q5_S-00004-of-00004.gguf filter=lfs diff=lfs merge=lfs -text
 imatrix.gguf filter=lfs diff=lfs merge=lfs -text
+Q4_K/MiniMax-M2.7-256x4.9B-Q4_K-00001-of-00004.gguf filter=lfs diff=lfs merge=lfs -text
+Q4_K/MiniMax-M2.7-256x4.9B-Q4_K-00002-of-00004.gguf filter=lfs diff=lfs merge=lfs -text
+Q4_K/MiniMax-M2.7-256x4.9B-Q4_K-00003-of-00004.gguf filter=lfs diff=lfs merge=lfs -text
+Q4_K/MiniMax-M2.7-256x4.9B-Q4_K-00004-of-00004.gguf filter=lfs diff=lfs merge=lfs -text

Q4_K/MiniMax-M2.7-256x4.9B-Q4_K-00001-of-00004.gguf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:9ba34e5363577d34bf56e655d603180f4ed42ffb994fd52f0a8ddee6fb724b43
+size 8237504

Q4_K/MiniMax-M2.7-256x4.9B-Q4_K-00002-of-00004.gguf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:6f959188e12f95f08c0907089e49fb29c12486dc95f33bcab8f8727c332792e7
+size 41983346304

Q4_K/MiniMax-M2.7-256x4.9B-Q4_K-00003-of-00004.gguf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:ceab2f20f0c2651f2738b9891e72979a0d06aa090c76a102ff3c49ce72cfc412
+size 41879006048

Q4_K/MiniMax-M2.7-256x4.9B-Q4_K-00004-of-00004.gguf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:e48e87054513246853b8ccdcfeb3d64d6a4d3f6a32b25c5298f6c563c0a333f7
+size 27032460384

README.md CHANGED Viewed

@@ -7,5 +7,4 @@ Minimax 2.7 quants where I tried to extract the maximum performance for my hardw
 This is the similar as AesSedai IQ4_XS quant (and uses the same imatrix) but the down tensors are replaced:
-- Q4_S replaces IQ4_XS with Q4_K
-- Q5_S replaces IQ4_XS with Q5_K


7
8	This is the similar as AesSedai IQ4_XS quant (and uses the same imatrix) but the down tensors are replaced:
9
10	+ - Q4_K replaces IQ4_XS with Q4_K

scripts/quantize.sh CHANGED Viewed

@@ -4,14 +4,14 @@ set -euo pipefail
 recipes=(
   "
-  MIX=Q5_S
   TYPE_FFN_GATE_UP_EXPS=IQ3_S
   TYPE_FFN_DOWN_EXPS=Q5_K
   TYPE_DEFAULT=Q8_0
   "
   "
-  MIX=Q4_S
   TYPE_FFN_GATE_UP_EXPS=IQ3_S
   TYPE_FFN_DOWN_EXPS=Q4_K
   TYPE_DEFAULT=Q8_0

 recipes=(
   "
+  MIX=Q5_K
   TYPE_FFN_GATE_UP_EXPS=IQ3_S
   TYPE_FFN_DOWN_EXPS=Q5_K
   TYPE_DEFAULT=Q8_0
   "
   "
+  MIX=Q4_K
   TYPE_FFN_GATE_UP_EXPS=IQ3_S
   TYPE_FFN_DOWN_EXPS=Q4_K
   TYPE_DEFAULT=Q8_0