Instructions to use hermitdave/Hemmingway-1-oQ4e with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hermitdave/Hemmingway-1-oQ4e 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("hermitdave/Hemmingway-1-oQ4e")

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 hermitdave/Hemmingway-1-oQ4e with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "hermitdave/Hemmingway-1-oQ4e"

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "hermitdave/Hemmingway-1-oQ4e"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use hermitdave/Hemmingway-1-oQ4e with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "hermitdave/Hemmingway-1-oQ4e"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "hermitdave/Hemmingway-1-oQ4e"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "hermitdave/Hemmingway-1-oQ4e",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use hermitdave/Hemmingway-1-oQ4e 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 "hermitdave/Hemmingway-1-oQ4e"

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 hermitdave/Hemmingway-1-oQ4e

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use hermitdave/Hemmingway-1-oQ4e with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "hermitdave/Hemmingway-1-oQ4e"

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 "hermitdave/Hemmingway-1-oQ4e" \
  --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"

Hemmingway-1 oQ4e (MLX)

oQ4e (enhanced/GPTQ) mixed-precision quantization of Altworld/Hemmingway-1, a 27B parameter hybrid-attention language model built on Qwen3.8-27B.

Upstream model: Altworld/Hemmingway-1 by Altworld, released under Apache-2.0.

Conversion: Quantized to MLX format using Hermes Agent with oMLX (oQ4e enhanced path with GPTQ weight optimization).

Architecture

Component	Value
Parameters	27B
Hidden size	5120
Layers	64
Attention heads	24 (full) / 16 (linear)
KV heads	4 (full) / 48 (linear)
Intermediate size	17408
Vocab size	248,320
Context length	262,144
Attention type	Hybrid (linear + full every 4th layer)
RoPE theta	10,000,000

Quantization Details

Property	Value
Method	oQ4e (enhanced/GPTQ)
Base bits	4
Target bpw	~4.6
Group size	64
Calibration	oqe_code_multilingual (128 samples, 512 seq length)
GPTQ	Yes (Hessian-based error compensation)
MTP	Stripped (preserve_mtp=False)
Dtype	bfloat16

Quickstart

pip install -U mlx-lm

python3 -m mlx_lm.generate \
  --model hermitdave/Hemmingway-1-oQ4e \
  --prompt "Write a text to your landlord about the broken boiler." \
  --max-tokens 512 --temp 1.0 --top-p 0.95

oMLX Inference

# Copy to oMLX models folder
cp -r ~/Projects/ml/models/Hemmingway-1-oQ4e ~/.omlx/models/hermitdave/

# Serve via oMLX
omlx serve --model hermitdave/Hemmingway-1-oQ4e

Benchmark Results (DeepEval, local run)

Head-to-head vs Ornith-1.5-35B oQ4e, measured on this quant with DeepEval GEval (LLM judge), run 2026-09-23, oMLX on M3 Max. 50 cases per model across 5 categories; message-not-memo sampled 3 generations per prompt.

Category	Hemmingway-1 oQ4e	Ornith-1.5 oQ4e
Directness	1.00	0.78
Human-Likeness	1.00	0.72
Empathy (EQ)	0.72	0.85
Hard asks — Actionability	0.64	0.53
Hard asks — Confidence	0.62	0.75
Message Not Memo	0.71	0.82
Story Quality	0.94	0.96

Story quality is on par with Ornith (0.94 vs 0.96) — an earlier run that showed a story gap used a 1024-token cap that truncated outputs before the message was written; with a 4096-token cap the gap disappears. ~5% of generations hit transient server errors and were excluded.

Citation

@misc{hemmingway1,
  title  = {Hemmingway-1},
  author = {Altworld},
  year   = {2026},
  url    = {https://huggingface.co/Altworld/Hemmingway-1},
}

License

Apache-2.0 (same as upstream).

Downloads last month: 559

Safetensors

Model size

27B params

Tensor type

U32

BF16

MLX

Hardware compatibility

4-bit

Model tree for hermitdave/Hemmingway-1-oQ4e

Base model

Qwen/Qwen3.8-27B

Finetuned

Altworld/Hemmingway-1

Quantized

(35)

this model

Collection including hermitdave/Hemmingway-1-oQ4e

oQ Quants

Collection

oQ qunats built with oMLX • 12 items • Updated about 21 hours ago