MLX 4-bit build of
Eikos-4B
: MLX weights quantized to 4 bits (group size 64;
mlx_lm convert -q --q-bits 4 --q-group-size 64
) for Apple Silicon.
The vision tower, MTP weights, embeddings and LM head are kept in higher precision. The prompt format, the
letter readout and the calibration (
calib.json
, T = 1) are the same as the bf16 model.
Eikos answers typed decisions (yes/no, one of N options, ordinal scores) about a given state in one forward pass,
with a calibrated probability for every option. See the
Eikos-4B card
for what the
model does, how it was trained, the full evaluation and its limitations.
Use
pip install mlx mlx-lm
hf download caiovicentino1/Eikos-4B-MLX-4bit --local-dir Eikos-4B-MLX-4bit
python Eikos-4B-MLX-4bit/mlx_decide.py Eikos-4B-MLX-4bit # runs a few demo decisions
import sys; sys.path.insert(0, "Eikos-4B-MLX-4bit")
from decision_core import options_of
from mlx_decide import MLXDecider
d = MLXDecider("Eikos-4B-MLX-4bit")
q = {"type": "noul", "instructions": "Under rule 4.2, can this order be executed as submitted?",
"criteria": {"true": "complies with rule 4.2", "false": "breaches rule 4.2"}}
probs, n_tokens = d.dist("Order ticket #A-2231 ... Rule 4.2: ... Approvals on file: none.", q, options_of(q))
mlx_decide.py
uses the same letter readout, prompt and calibration as the GPU builds. It also answers several
questions over one state in a single cached batch (
dist_many_cached
).
Validation against bf16
Same 7,371 items for both builds (7 suites, never used in training), MLX run with its CUDA backend on Linux.
The release gate was fixed before looking at results: accuracy within 1 point of bf16, ECE within 0.01, and at
least 97% of answers unchanged.
Eikos-4B (bf16)
Eikos-4B-MLX-4bit
Size
9.3 GB
2.4 GB
JevBench public — original / hard
91.7 / 73.9
95.8 / 73.0
DecisionBench — medium / hard
77.1 / 66.6
78.8 / 64.8
General battery (9 tasks)
75.7
75.3
Finance (CUAD, sentiment, FinQA-judge)
74.7
76.0
Trade rules — seen / unseen
74.5 / 76.0
72.4 / 76.0
Compositional rules — same type / new domain / rulebooks
96.0 / 91.5 / 91.7
95.0 / 90.5 / 90.7
ECE (lower is better)
0.033
0.021
≥0.90 confidence: decides / error
34.7% / 2.5%
32.5% / 2.7%
Same answer as bf16 (all / confident ≥0.9)
—
92.5% / 99.8%
It passes our release gate on accuracy and calibration, but not on answer agreement (92.5% of answers are the same as bf16; the gate asks for 97%). Most changed answers (93%) are on items where the bf16 model itself was unsure (confidence below 0.7); on decisions the bf16 model takes with confidence ≥0.9, 99.8% of answers are the same.
License
MIT for our contributions (
LICENSE
). The base model, Qwen3.5-4B, is Apache-2.0 (
LICENSE-Qwen
);
attributions are in
NOTICE
. Not legal, tax or investment advice.
Runs of caiovicentino1 Eikos-4B-MLX-4bit on huggingface.co
88
Total runs
0
24-hour runs
1
3-day runs
9
7-day runs
87
30-day runs
More Information About Eikos-4B-MLX-4bit huggingface.co Model
Eikos-4B-MLX-4bit huggingface.co is an AI model on huggingface.co that provides Eikos-4B-MLX-4bit's model effect (), which can be used instantly with this caiovicentino1 Eikos-4B-MLX-4bit model. huggingface.co supports a free trial of the Eikos-4B-MLX-4bit model, and also provides paid use of the Eikos-4B-MLX-4bit. Support call Eikos-4B-MLX-4bit model through api, including Node.js, Python, http.
Eikos-4B-MLX-4bit huggingface.co is an online trial and call api platform, which integrates Eikos-4B-MLX-4bit's modeling effects, including api services, and provides a free online trial of Eikos-4B-MLX-4bit, you can try Eikos-4B-MLX-4bit online for free by clicking the link below.
caiovicentino1 Eikos-4B-MLX-4bit online free url in huggingface.co:
Eikos-4B-MLX-4bit is an open source model from GitHub that offers a free installation service, and any user can find Eikos-4B-MLX-4bit on GitHub to install. At the same time, huggingface.co provides the effect of Eikos-4B-MLX-4bit install, users can directly use Eikos-4B-MLX-4bit installed effect in huggingface.co for debugging and trial. It also supports api for free installation.