Downloads · 30 days
40
26% of all-time downloads
morgan/qwen38-27b-mtp-r20k-lr3-q4-g64-q2-rerank
qwen38-27b-mtp-r20k-lr3-q4-g64-q2-rerank is a machine learning model from morgan. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for mlx. The card lists the license as apache-2.0.
This repository contains a proposal-head artifact for native-MTP speculative decoding with Qwen 3.8 27B. It is not a standalone language model. The fixed target backbone and tokenizer are intentionally not included.
Downloads · 30 days
40
26% of all-time downloads
All-time downloads
156
Public
Parameters
129M
428 MB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors428 MB · 100%
How the weights are stored.
U3284.6M · 65%
From the Hugging Face model README
This repository contains a proposal-head artifact for native-MTP speculative decoding with Qwen 3.8 27B. It is not a standalone language model. The fixed target backbone and tokenizer are intentionally not included.
The artifact composes the LR3 trained MTP head with the public promoted Qwen 3.8 challenge head ABI. The composition changes proposal values while preserving the promoted runtime mechanisms:
draft_lm_head weight, scales, and biases are copied
byte-for-byte. The compatible runtime uses this triple only to build a
32-token shortlist, then performs the promoted exact affine-4 rerank.The Q-row selector is intentionally not recomputed. Preserving the public frontier layout holds the runtime and shortlist mechanism fixed so the Apple comparison isolates the trained core and gathered island values.
The head only proposes future tokens. The fixed target model still verifies every emitted token, and the trusted parent still owns exact target top-two evidence, row accounting, rollback, replay, and scoring. The artifact contains no executable remote code.
| Field | Value |
|---|---|
model.safetensors bytes | 427742680 |
| Raw file SHA-256 | 5aa88cfc87a5a14cc29bf098a4d049ff537756618af86f38f0364d5c87d07d3d |
| Model-only tree SHA-256 | 2f6805e1c8b7ea5e5c6cf19acc968426c5b1c3fb2e5442eb1000df96550bf174 |
| Tensor count | 40 |
| Tensor payload bytes | 427738112 |
config.json bytes | 3570 |
config.json SHA-256 | fb2a5bd03bc4a2ac6a161f2f2eb5bf9fc3e4e23bf50a55c2ba41fe6b692d2b95 |
| Fine-tuned BF16 source bytes | 849400336 |
| Fine-tuned BF16 source SHA-256 | 277b14f2e81d71beb4824540c1afc9aa2dfdbecfef0ab3469feb28a7511a3714 |
| Packed LR3 Q4/G64 SHA-256 | aefad31ef49f45cdfe4587d36f6f20b97bc0dd1e0b45bef0587d403f48e6c32c |
The model-only tree digest is SHA-256 over this exact line, including its final newline:
5aa88cfc87a5a14cc29bf098a4d049ff537756618af86f38f0364d5c87d07d3d model.safetensors
The source frontier head is
amal-david/qwen38-mtp-head-q2-q4-rerank-v1@ae6282749a52e052496dd5300b4aa441df7301e8.
Its raw model SHA-256 is
d038fd41e2d5dab1b3905c115d859fdc98dfbfde9862c14ebb82c2b3247ec2f1,
and its model-only tree SHA-256 is
559b24ebca354018e4402fdb1f5af1afe5a0721bd2ebf04133500d846f7d5f71.
The LR3 compositor verified that the q2 readout triple, all three precision- island index arrays, and tensor wire order are byte-exact copies of that frontier artifact. It also verified that all 31 packed core tensors match the LR3 Q4/G64 build and every island row matches its LR3 BF16 source row.
LR3 was trained from source commit
79cce6cb594aefc53294cdb0aabb277997ac863d. It used a K=4 objective with
acceptance-loss beta 0.6, a 4,000-step cosine horizon, 200 warmup steps,
AdamW learning rate 3e-6, and Muon learning rate 3e-5. The selected
checkpoint was step 2,469 and was also the final checkpoint. Its validation
objective was 0.5906565.
The source is the 94,562-row safe Dev20 tree with SHA-256
85a444fe8476085ac92f1f110e0b01aca33cc98b74758ea2019910d121942130.
Training uses the 20,000 rows with the smallest
BLAKE2b-128(seed:source_index) keys. The selected-row tree SHA-256 is
d17ed48d18d0c4b260bec9c478fcab98c54c9104df0cbad02284cc68a8703211,
and the exact membership-manifest SHA-256 is
57c62cb68d4ede120698acdc1ab53d6dcf0e9ac781e2b716b6f4ffca76c8fff1.
The prepared cache contains those 20,000 rows plus 4,049 held-out validation
rows, for 24,049 rows total.
The CUDA directional diagnostic measured mean committed length
3.4625480154 and accepted-draft rate 0.6156370. It matched the CUDA serial
stream on 96 of 256 samples. That exactness result is diagnostic-only because
the CUDA Qwen-GDN path is numerically non-invariant; Apple MLX remains the
deployment correctness authority.
The matched control measured committed length 3.4360958033064857, accepted-
draft rate 0.6090239508266214, and per-depth acceptance
0.8256676557863502 / 0.6611911827045358 / 0.5277660025434506 / 0.4214709622721492. LR3 improved committed length by
0.0264522120584313, or about 0.7698%, in this directional diagnostic.
LR3 per-depth acceptance was 0.832266325 / 0.668053777 / 0.533504055 / 0.428723858.
The public W&B run is
q38mtp-20260818-073519-7afdfb49.
The LR3, B05, and matched-control W&B runs are terminal finished.
The terminal deployment screen ran on an Apple M4 Pro Mac mini with 48 GiB of
memory, macOS 26.5.2 (25F84), Xcode 26.6 (17F113), Swift 6.3.3, and Metal
toolchain com.apple.dt.toolchain.Metal.32023.883. It used exact campaign base
a187ec6c479bfec7e62fbb55a68c39aa128ceca7, a 128-token window, MTP depth 8,
natural cooling, fan auto, and three real GPU cool gates at or below 40C per
arm. These are directional local results, not official ranked scores.
The first sequence used order control/LR3/LR3/control:
| Arm | Head | MTP seconds/token |
|---|---|---|
| 01 | Current frontier | 0.05670436006039381 |
| 02 | LR3 | 0.055505343712866306 |
| 03 | LR3 | 0.055216390639543533 |
| 04 | Current frontier | 0.056804000400006771 |
The control mean was 0.0567541802302003 seconds/token and the LR3 mean was
0.0553608671762049. LR3 was 2.454996% faster by the matched absolute-time
criterion.
The second sequence used order LR3/B05/B05/LR3:
| Arm | Head | MTP seconds/token |
|---|---|---|
| 01 | LR3 | 0.055070945061743259 |
| 02 | B05 | 0.055201351642608643 |
| 03 | B05 | 0.055350874550640583 |
| 04 | LR3 | 0.055238242261111736 |
The LR3 mean was 0.0551545936614275 seconds/token and the B05 mean was
0.0552761130966246. LR3 was 0.219841% faster, which is inside the
predeclared ±0.5% noise band. LR3 was selected because it decisively beat the
current frontier and did not lose the B05 comparison.
Every arm emitted all 128 expected tokens, reported zero residual divergence,
and closed its reference-row ledger. LR3 and B05 each had 17 rounds, effective
mean draft length 6.5294117647058822, and accepted-draft rate 1.0 on this
public prompt. The current-head control had 20 rounds, effective mean draft
length 5.7, and accepted-draft rate 0.956140350877193.
See provenance.json for the machine-readable record. Any consuming manifest
must pin an immutable Hugging Face revision and independently reproduce the
declared artifact identity before use.