Instructions to use davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw")
model = AutoModelForCausalLM.from_pretrained("davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Trellis

How to use davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw with Trellis:

# No code snippets available yet for this library.

# To use this model, check the repository files and the library's documentation.

# Want to help? PRs adding snippets are welcome at:
# https://github.com/huggingface/huggingface.js

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw

SGLang

How to use davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw with Docker Model Runner:
```
docker model run hf.co/davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw
```

GLM-5.3 — EXL3/TR3 3.25 bpw (mixed K3/K4 trellis, data-free)

Weights uploaded and verified; KLD measured (2026-08-29, full-vocabulary teacher-forced KLD vs the sealed BF16 teacher, held-out confirmation windows of brandonmusic/GLM-5.3-BF16-full-logits — 4 windows x 2,047 positions x 154,880 vocab; reproduction kit + runner in this repo).

Trellis (EXL3) quantization of zai-org/GLM-5.3 (755B glm_moe_dsa MoE: 78 layers + MTP, 256 routed experts/layer), following the GLM-5.2 TR3 lineage layout. Fits TP4 on 4x 96 GB (RTX PRO 6000 Blackwell class) with FP8 KV cache.

From the upstream card: "GLM-5.3 uses the same base model as GLM-5.2 — every gain comes from post-training. Compared with GLM-5.2, it is much better at complex coding and long-horizon tasks." See the base model card for benchmarks, serving guides, and the technical report (arXiv:2602.15763).

What is quantized

component	treatment
routed experts (layers 3–78, incl. MTP-78)	EXL3 trellis, per layer 192 experts K3 + 64 experts K4 (avg 3.25 bpw), mcg codebook
dense MLP (layers 0–2), all attention, norms, embeddings, lm_head, mlp.gate, eh_proj	BF16, carried byte-exact
shared experts	BF16 in-checkpoint (online K6 at serve)

Data-free encode: identity Hessian (H=I), rotations + trellis search only — no calibration capture anywhere. Deterministic: seeds derived from (layer, expert, projection, rank), seed_base 20260711.
K4 selection: per layer, the experts with highest relative round-trip MSE under the K3 encode are re-encoded at K4 (worst-64 per layer).
Per-expert tier map in tier_bitmap.json; encode provenance in config.json.hybrid_tr3_tail; file hashes in MANIFEST.sha256.

Serving

TR3 mixed-bit layout (carrier BF16 shards + per-layer trellis payloads) — serve with an exllamav3-b12x/sparkinfer-lineage stack, TP4 (+ DCP4/MTP3), FP8 KV. The mixed-K projection-tiers patch is REQUIRED; a stock loader that assumes a uniform K per layer will produce fluent garbage. Not loadable by vanilla exllamav3 model loading.

KLD vs BF16 teacher

Full-vocabulary, teacher-forced KL(teacher || student) against sealed BF16 GLM-5.3 logits — 4 held-out windows × 2,047 positions × 154,880 vocabulary, fp32 log-softmax both sides. Measured independently on two different 4× RTX PRO 6000 (96GB) machines:

weight quant	KV mode	this work	CN3 (@dareposte)	Δ
3.42 bpw	fp8	0.024105	0.023966	−0.6%
3.25 bpw	fp8	0.026103	0.026776	+2.6%
3.25 bpw	nvfp4	0.035741	0.036661	+2.6%
3.42 bpw	nvfp4	0.037757	0.037060	−1.8%
3.42 bpw	nvfp4+rope8	0.039518	0.037695	−4.6%
3.25 bpw	nvfp4+rope8	—	0.039396	CN3 only

Readings: the 3.25↔3.42 weight step changes KLD by only ~~0.002; the fp8→nvfp4 KV step costs ~7× more (~~0.014) — cache format matters more than the extra 0.17 bpw. Under nvfp4 KV the weight-quant difference washes out entirely. The window SD (~0.02–0.03) is corpus heterogeneity — one citation-dense legal window is uniformly hardest; dialogue, explanatory prose, and reasoning-trace registers measure near-transparent.

Per-window means (dialogue / legal / prose / reasoning-trace)

config	source	w0000	w0001	w0002	w0003
3.42 fp8	this work	0.0141	0.0537	0.0137	0.0148
3.42 fp8	CN3	0.0148	0.0542	0.0135	0.0134
3.25 fp8	this work	0.0188	0.0579	0.0138	0.0139
3.25 fp8	CN3	0.0183	0.0592	0.0142	0.0154
3.42 nvfp4	this work	0.0199	0.0816	0.0221	0.0274
3.42 nvfp4	CN3	0.0198	0.0775	0.0210	0.0300
3.25 nvfp4	this work	0.0256	0.0799	0.0184	0.0190
3.25 nvfp4	CN3	0.0255	0.0804	0.0189	0.0218
3.42 nvfp4+rope8	this work	0.0212	0.0812	0.0257	0.0300
3.42 nvfp4+rope8	CN3	0.0218	0.0761	0.0236	0.0292
3.25 nvfp4+rope8	CN3	0.0268	0.0883	0.0204	0.0221

Method (reproducible)

Teacher: brandonmusic/GLM-5.3-BF16-full-logits, reference-full-panel confirmation lane (held out from every calibration fit), revision 427368f1.
Student: this checkpoint, loaded by the digest-pinned r17 serving image (sha256:c5e96c5b…) — the real trellis kernels and online-K6 path, offline vllm.LLM, TP4, one teacher-forced prefill per window.
Full runbook + runner: kld/ in this repo (KLD-REPRODUCTION.md, prefill_kld_53.py, fetch-teacher.sh).
Independent reproduction bundle (receipts, unedited logs, checksums, pinned revisions): kld/cn3/.

Capacity caveat (via the CN3 report): KV-pool token figures printed by KLD-profile boots (TP4/DCP1, 4,096-token envelope) are logical pool values for that profile only — they are not maximum context length or production serving capacity.

Credits

brandonmusic — thank you for the GLM-5.3-BF16-full-logits teacher captures that make this measurement possible without a 1.5TB BF16 forward, and for the TR3 quantization references this release follows: the GLM-5.2-EXL3-TR3v4-3.5bpw-MTP78 method card, runbook, and the r10 reproducibility bundle whose encoder lineage (encode_tr3_v31.py) this checkpoint was produced with.
dareposte — thank you for the independent CN3 reproduction: all six weight/KV configurations on separate hardware, within ±5% of our means, published here with receipts under kld/cn3/.
willfalco/GLM-5.2-EXL3-TR3-3.25bpw — the 3.25 tier recipe and mixed-K checkpoint-format lineage.
local-inference-lab — the qualified r17 serving stack these artifacts boot on.

Status

Measured and independently reproduced 2026-08-29: mean KLD 0.026103 (fp8 KV; CN3 reproduction 0.026776) on held-out confirmation windows. Serving-qualified on the r17 stack; measured 590k-token KV pool at GMU 0.965 — the context-breadth artifact of this release pair.

Quantized with encode_tr3_53.py (exllamav3 v0.0.43 vendored math, MIT) on 4x RTX PRO 6000 Blackwell. Encode + tooling notes ship in the repo.

Downloads last month: 801

Safetensors

Model size

170B params

Tensor type

BF16

F32

F16

I16

Model tree for davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw

Base model

zai-org/GLM-5.3

Quantized

(28)

this model

Paper for davidsyoung/GLM-5.3-EXL3-TR3-3.25bpw

GLM-5: from Vibe Coding to Agentic Engineering

Paper • 2602.15763 • Published Feb 17 • 217