Downloads · 30 days
3
27% of all-time downloads
brandonmusic/GLM-5.2-SQG-Coupled-H512-H128-K96Tail
GLM-5.2-SQG-Coupled-H512-H128-K96Tail is a text generation model from brandonmusic. Use it when you need the model to write or continue text. It is set up for vllm. The card lists the license as other.
Model status: research-only. The checkpoint is mechanically complete and its TP4/DCP4/MTP3 runtime is qualified, but full-vocabulary KLD is 84.849 percent worse than the frozen SQG source. The public file set is unsup…
Downloads · 30 days
3
27% of all-time downloads
All-time downloads
11
Public
Repo size
144 GB
Likes
0
Public
Click a slice to open those files.
.safetensors143 GB · 99%
From the Hugging Face model README
Model status: research-only. The checkpoint is mechanically complete and its TP4/DCP4/MTP3 runtime is qualified, but full-vocabulary KLD is 84.849 percent worse than the frozen SQG source. The public file set is unsupported as a complete model until routed layers 3 through 50 finish uploading and a hash-bound revision passes anonymous verification.
This checkpoint re-encodes the routed experts from source SQG-W4A8 revision using saved calibration captures and Hessians from Hessian dataset revision. The encode did not download or read the official BF16 routed weight shards. The KLD methods did use BF16 reference logits and the unchanged BF16 LM head; that evaluation use is not an official BF16 routed-weight read.
| Object | Status | Result |
|---|---|---|
| Coupled encoding implementation | implemented | Residual H512, H128 before and after the activation, exact silu(gate) * up, H13 alpha 0.25, and candidate-conditioned H2 are bound in the receipts. |
| Routed layers 3 through 77 | qualified | All 75 target layers have sealed manifests, quality receipts, and passing TP1 native runtime oracles. Exact K96 scorer/encoder parity covers layers 4 through 77. Layer 3 is the sealed K48 exception and has no K96 parity receipt. |
| Local assembly and codec census | qualified | All 76 routed layers pass, including preserved source MTP layer 78. |
| Exact-r11 TP4/DCP4/MTP3 runtime | qualified | Four rank receipts, loaded and executed layers 3 through 78, no fatal audit matches, and sealed MTP3 metrics. |
| Estonia and LAVD task runs | qualified | Estonia is 5/5 correct. LAVD is 4 exact plus 1 near under the published tolerance. |
| Full-model KLD quality | research-only | Candidate mean KLD 0.1401771516114036 is worse than source mean 0.07583317451217256. |
| Hidden replay | research-only | The original preregistered maximum-position limit fails. A post-observation operational envelope passes. |
| Public file set | unsupported | Routed layers 3 through 50 and final anonymous verification are incomplete. |
implemented, qualified, research-only, and unsupported are the only
status labels used by this card. Task behavior and serving closure do not
override the failed full-model KLD gate.
The checkpoint uses the updated-QSRT coupled transform. Each routed expert is treated as one gate/up/down function:
silu(gate) * up;0.25; andLayers 4 through 77 run the full per-layer profile and beta search. Fleet beta reuse, the B300 owner-speed rescue, identity-only fallback, and byte-changing same-rate batching are excluded.
| Layer range | Treatment | K3 | K4 | Routed bpw |
|---|---|---|---|---|
| 3 | Sealed coupled K48 layer | 720 | 48 | 3.0625 |
| 4 through 77 | Coupled K96 layers | 672 | 96 | 3.125 |
| 78 | Preserved source MTP layer | 384 | 384 | 3.5 |
Target layers 3 through 77 average 3.1241666666666665 bpw. All 76 routed
layers, including MTP layer 78, average 3.1291118421052633 bpw. This model is
not uniform 3.0625 bpw and is not uniform K96.
Non-routed and BF16 components remain byte-identical to the frozen source. MTP layer 78 is also retained from the source.
The primary quality measurement is KL(BF16 reference || candidate) over all
154,880 vocabulary entries at each of 2,047 causal positions. It uses the same
fixed 2,048-token input as the source measurement, no position trimming, and
TP4/PP1/DCP1.
| Metric | Frozen source | Candidate |
|---|---|---|
| Mean | 0.07583317451217256 | 0.1401771516114036 |
| Median | 0.0015440876595675945 | |
| p95 | 0.6778242588043213 | |
| p99 | 1.397577404976 | 2.480538845062256 |
| Worst-1% CVaR | 2.207112874304 | 4.160942645300002 |
| Maximum | 5.978030681610 | 8.928885459899902 |
| Finite positions | 2047/2047 | 2047/2047 |
The candidate mean is 0.06434397709923104 higher, 1.84849x the source
mean, and 84.849 percent worse. The required lower mean, lower p99, and lower
worst-1% CVaR gates fail. This is why the model is research-only.
Candidate receipt SHA-256:
7979c9c8b0c81714cd38e225646e42a88b2cb8eb03232be271373255c506a408.
Source receipt SHA-256:
5d8aedb462658c693f1ce790f48ce5ed3cd6876897b1367a5cd36e42c0e2d434.
The operational hidden replay uses a raw BF16 [2048,6144] pre-LM-head
capture and the unchanged BF16 [154880,6144] LM head. It reports mean KLD
0.1401762649458023 and top-1 agreement 0.9174401563263312 across 2,047
positions. Operational receipt SHA-256:
34f14cada0424ddb1387fec78a96a16ebe2109ddf2c063862d52108d0450e6b2.
The original preregistered receipt has SHA-256
7433ad312740dabaa1f3dc0c6e2a8317e741a441bccf06eb4a6f2ebe0569b3ad
and qualification_pass=false. Mean absolute KLD delta is
8.866656012740393e-7, below the 5e-5 limit, but maximum position delta is
about 0.0035558, above the preregistered 5e-4 limit.
A later operational envelope uses mean <=5e-5, p99 <=1e-4, and maximum
<=5e-3; it passes. Those limits were chosen after the discrepancy was
observed, so that pass is research-only and is not preregistered qualification.
Original comparison script SHA-256:
655301634275b617cb7e933a698811ad0605941c8d31264b4c2410086c05a038.
Original service journal SHA-256:
380f3ed2e6c7ee8c46953a1fb3ab48678cdde813ceda489a348c7e8f6845c939.
The qualified runtime is derived from Infernal Invocation r11. Infernal Invocation r13 supplies native-SQG donor code only. v20 is not part of the runtime lineage.
verdictai/glm52-k96-ii-r11:20260815-tpfix-mtpfixsha256:ab6bd60716b0a8e453b6345cb10e43e79726d92729b1f29058e31d7cc1c67defnvfp4_ds_mla4f19ba5e4a8676c80bc49e89d346b0985faa209f14bdd6d9713e9ee6c4397f57d8a7f22f6da05423a00d972e186deb17ba9f36133aae692c2477d53cd4f0f4ffThe qualification bundle's SHA256SUMS validates all sealed files.
Five of five runs are correct, with zero errors and zero truncations. Average
completion length is 3,371.6 tokens. Average generation speed per run is
43.5170233617816 tokens per second.
Five of five runs are correct under the published tolerance: 4 exact and 1
near, with zero errors and zero truncations. The near answer is 71, 45.75
against expected 72, 46. Average completion length is 16,682.2 tokens.
Average generation speed per run is 22.087665860284208 tokens per second.
This result must not be described as 5 exact.
The runtime generated 106,204 tokens, drafted 88,704 tokens, and accepted
76,631 draft tokens. Aggregate acceptance is 0.863895652958153. Acceptance
by speculative position is 0.9242762445887446, 0.8623511904761905, and
0.8050595238095238.
The MTP EXL3 patch maps the speculative mtp_block module prefix to the
canonical layer-78 checkpoint prefix. A real-image static check passed for both
prefix forms. Full pytest did not run because a dependency was unavailable, so
no full-pytest result is claimed.
The browsable reproduction closure contains:
The human procedure is
reproduction/docs/K96_COUPLED_DISTRIBUTED_REPRODUCTION_20260814.md. The
machine contract is
reproduction/machine/k96tail-distributed-campaign.json.
Server-side copy commit
0c38e683eda27ca84982e3d513c89dd780dcdb22 copied and verified 390
byte-identical files totaling 25,685,857,224 bytes from the frozen source
revision. Routed layers 51 through 77 were already present, and MTP layer 78
was copied. The active targeted upload restarted at
2026-08-15T05:39:16-04:00 with one outer client and default adaptive Xet.
Adaptive concurrency began at 2. Early sustained evidence showed 22.09 Mbit/s,
success ratio 1.0, and zero errors. No completion ETA is part of this card.
The repository is unsupported as a complete downloadable model until all missing layers finish, the canonical full-folder verification passes, this research-only quality disclosure is published, and a hash-bound public revision passes anonymous representative-file hash checks.
The follow-on upload does not wait for the intentionally stopped qualification
server. It excludes this README.md and HUB_FILE_VERIFICATION.json, uploads
every remaining file with two workers, and verifies the resolved Hub revision
against local bytes. LFS files must match SHA-256 and size; Git files must
match Git blob SHA-1 and size. The verification receipt is uploaded only after
every file matches. No passing public verification receipt exists at this
documentation snapshot.
3acea57fb5a82b3b848504bdc87953420f531f39
through audited snapshot 104dd9233f850a3955f4991bea68b07dd34deeb8 is
authored by Luke. The repository is not linked because no working public URL
was verified.This derivative does not relicense GLM-5.2 or third-party runtime/code. It remains subject to the upstream GLM-5.2 license and the licenses of the linked source projects. Reused ExLlamaV3 components remain subject to its MIT license. The audited QSRT/KQuant snapshot contains no top-level license declaration, so this card does not infer one; resolve that source-license notice before redistributing those source files independently.