h

h

Same weights as hermitdave/Hemmingway-1-MLX-4bit. The tokenizer strings for the chat and thinking markers are one character, and they still encode to the original token IDs.

Uniform 4-bit 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 mlx-lm (uniform 4-bit, group size 64).

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 Uniform 4-bit
Bits per weight 4.501
Group size 64
Dtype bfloat16

Tokenizer markers

These replace the long control strings. Token IDs are unchanged, so the weights are not retrained.

Token ID Was Now
248045 <|im_start|> ≺
248046 <|im_end|> ≻
248068 <think> ⊏
248069 </think> ⊐

A prompt that used to start <|im_start|>system is now ≺system. The old long strings are ordinary text and do not map to those IDs anymore.

Quickstart

pip install -U mlx-lm

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

Benchmark Results (DeepEval, local run)

Benchmarked on the oQ4e quant of this model (not this 4-bit quant) in a head-to-head vs Ornith-1.5-35B oQ4e, DeepEval GEval (LLM judge), run 2026-09-23, oMLX on M3 Max. 50 cases per model; 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

~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
103
Safetensors
Model size
27B params
Tensor type
U32
·
BF16
·
MLX
Hardware compatibility
Log In to add your hardware

4-bit

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

Model tree for hectoreccles/h

Base model

Qwen/Qwen3.8-27B
Quantized
(1)
this model