GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
kvit-s
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
kvit-s
Model Tree
Cloudflare/clef
Input Modalities
Output Modalities
Supported Functionality
| File | What it is |
|---|---|
adapter_model.safetensors | 186 pairs of LoRA matrices, rank 256, bf16; 700,366,848 parameters, 1.40 GB |
adapter_config.json | PEFT configuration: r = lora_alpha = 256, layers 40–63, twelve module types (q_proj, k_proj, v_proj, o_proj, in_proj_qkv, in_proj_z, in_proj_a, in_proj_b, out_proj, gate_proj, up_proj, down_proj) |
extraction-report.json | per-matrix figures from the extraction (below) |
reference/answers/ | Clef's answers at full precision (bf16, on the CPU) to the 72 test records of the GitHub repository |
reference/hidden/ | the hidden states of that run, to compare a server's output against (eval/compare_hidden.py) |
From Cloudflare/clef at revision 2f3de3dd85f379784083b0814d997ab627200f0c and
Qwen/Qwen3.8-27B at revision 1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0:
extract/extract_adapter.py in the GitHub repository repeats this with one command.
Clef's weights are stored in bf16, so each is already off by up to half a step of 8-bit precision. For the 150 matrices larger than 256 in both dimensions, what the factorisation leaves out measures 0.85 to 1.005 times that rounding noise, and for the smaller matrices a fraction of it. Weights rebuilt as base + B·A differ from Clef's by 0.17% at most.
Applied to a quantized base model, the adapter reproduces Clef less closely. On one RTX 5090 setup (a 4-bit NVFP4 checkpoint with a 4-bit cache, in vLLM) it answered 83 to 85 of 85 test questions with known answers correctly across runs, where Clef at full precision answered 83, and the median decision took 165 ms. The GitHub repository has the details and the scripts to measure other setups.
bash
export CLEF_LORA_LAYERS=40-63 # read by the clef_capture pluginvllm serve Qwen/Qwen3.8-27B ... \ --enable-lora --max-lora-rank 256 --max-loras 1 \ --lora-modules clef=kvit-s/clef-lora-r256Requests to the model name clef then run Clef's backbone; the plugin and the decision service
from the GitHub repository turn them into decisions.
Apache-2.0. The adapter is derived from the weights of Cloudflare's Clef and of Qwen3.8-27B, both released under Apache-2.0, as the difference between the two in decoder layers 40–63, factored into rank-256 matrices. No other change was made. The reference answers were computed with Cloudflare's code on rows of ARC-Challenge (allenai/ai2_arc, CC BY-SA 4.0) and BANKING77 (PolyAI, CC BY 4.0). The answer files hold record ids, option names, logits and probabilities. The hidden-state files also hold each prompt's token ids, which decode back to the prompt text, so for those records they contain text from the two datasets, under the datasets' licences.
export CLEF_LORA_LAYERS=40-63 # read by the clef_capture pluginvllm serve Qwen/Qwen3.8-27B ... \ --enable-lora --max-lora-rank 256 --max-loras 1 \ --lora-modules clef=kvit-s/clef-lora-r256