Instructions to use solarpush/push-ministral-8b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use solarpush/push-ministral-8b with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf solarpush/push-ministral-8b
# Run inference directly in the terminal:
llama cli -hf solarpush/push-ministral-8b

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf solarpush/push-ministral-8b
# Run inference directly in the terminal:
llama cli -hf solarpush/push-ministral-8b

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 solarpush/push-ministral-8b
# Run inference directly in the terminal:
./llama-cli -hf solarpush/push-ministral-8b

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 solarpush/push-ministral-8b
# Run inference directly in the terminal:
./build/bin/llama-cli -hf solarpush/push-ministral-8b

Use Docker

docker model run hf.co/solarpush/push-ministral-8b

LM Studio
Jan
Ollama
How to use solarpush/push-ministral-8b with Ollama:
```
ollama run hf.co/solarpush/push-ministral-8b
```
Unsloth Desktop

How to use solarpush/push-ministral-8b with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf solarpush/push-ministral-8b

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "solarpush/push-ministral-8b"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use solarpush/push-ministral-8b with Docker Model Runner:
```
docker model run hf.co/solarpush/push-ministral-8b
```

Lemonade

How to use solarpush/push-ministral-8b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull solarpush/push-ministral-8b

Run and chat with the model

lemonade run user.push-ministral-8b-{{QUANT_TAG}}

List all available models

lemonade list

Hermes Agent

How to use solarpush/push-ministral-8b with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf solarpush/push-ministral-8b

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 solarpush/push-ministral-8b

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use solarpush/push-ministral-8b with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf solarpush/push-ministral-8b

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 "solarpush/push-ministral-8b" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

push_ministral:8b — Ollaya & Ollama Calibrated Decision Model (GGUF Q4_K_M)

push_ministral:8b is a calibrated llm-logits-v1 structured decision model adapted from Ministral-8B-Instruct-2410 (Q4_K_M GGUF) using Trainaya's Biorthogonal Dual Template Basis (U / U_s) + Fisher ARD Kernel Option-Pointer Head.

Ollaya Public Decision Benchmark (`bench.ollaya.dev`, 13 Subsets)

Metric	`push_ministral:8b` (4.57 GiB)	`winnow:e4b` Baseline (6.20 GiB)
Macro Accuracy	69.1%	72.2%
Public Skill (0–100)	52.2	56.8
Accuracy (`Choice` / `Noul` / `Score`)	71.2% / 74.0% / 57.3%	76.8% / 76.0% / 58.0%
Pooled ECE (lower is better)	0.046	0.083
Macro ECE (lower is better)	0.125	0.133
Brier Score (lower is better)	0.434	0.393
Latency p50 (RX 9070 XT)	94 ms	113 ms

Highlights

Superior Probabilistic Calibration: Achieves 0.046 Pooled ECE and 0.125 Macro ECE (outperforming winnow:e4b).
Subset Wins / Ties vs winnow:e4b:
- civil_comments: 92.0% vs 78.0% (+14.0%, ECE 0.095)
- pubmedqa: 68.0% vs 62.0% (+6.0%, ECE 0.094)
- squad2: 66.0% vs 64.0% (+2.0%, ECE 0.082)
- summeval-consistency: 78.0% vs 76.0% (+2.0%)
- boolq: 80.0% vs 80.0% (+0.0%, ECE 0.066)

Bundle Contents

model.gguf — Q4_K_M GGUF with calibrated Q6_K (ttype=14) decision head rows (A..Z, 0..9).
decision.json — Ollaya llm-logits-v1 decision specification (choice, score, noul token IDs & template).
calibration.json — Per-task temperature scaling vector ([1.0, 1.0, 1.0]).
ollaya_config.json — Ollaya runtime metadata.

Usage with Trainaya (Ollaya + Ollama)

trainaya register --name push_ministral:8b --path . --runtime both

Downloads last month: -

GGUF

Model size

8B params

Architecture

llama

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for solarpush/push-ministral-8b

Base model

mistralai/Ministral-8B-Instruct-2410

Quantized

(71)

this model

push_ministral:8b — Ollaya & Ollama Calibrated Decision Model (GGUF Q4_K_M)

Ollaya Public Decision Benchmark (bench.ollaya.dev, 13 Subsets)

Highlights

Bundle Contents

Usage with Trainaya (Ollaya + Ollama)

Model tree for solarpush/push-ministral-8b

Ollaya Public Decision Benchmark (`bench.ollaya.dev`, 13 Subsets)