autotrust/GLM5.3-Flash-E256-DGX-Spark
image-text-to-text model by autotrust
Older snapshot: check the original source for current information.
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
[](https://news.guru.so/ui) autotrust/GLM5.3-Flash-E256-DGX-Spark is a compact build of zai-org/GLM-5.3-Flash for Blackwell systems, from DGX Spark clusters to a single B200. It's an unofficial derivative. It keeps 256 of the 288 routed experts in each layer by Neural Architecture Search (NAS), uses NVFP4 for the experts and still activates 18 B parameters per token. It's the largest architecture in the family, with the best Chinese, code and vision scores. The KV cache uses FP8. The repo includes a ready-to-use NVFP4 MTP draft in mtp-nvfp4/ (6.95 GB) for speculative decoding. On a single B200 it gives 1.84× single-stream decode speed (1.94× with 3 draft tokens), and output quality is unchanged. The model keeps the full 154,880-token vocabulary and has the vision tower intact. Measured on one B200 with vLLM. Sampling follows the base model's recipe (temperature=1.0, topp=0.95); HumanEval uses greedy decoding. reasoningeffort is the GLM-5.3-Flash chat-template thinking budget (low / high / max).…
Read the full model card ↗ · Preview checked 2026-10-30T04:36:29.716Z
Inside the original model card — Document outline
- GLM5.3-Flash-E256-DGX-Spark
- 🔴 Live: AutoTrust/GuruSearch recommendation & search demo → news.guru.so
- Benchmarks
- Tool use: BFCL v4 (function calling, AST match)
- Throughput and MTP speculative decoding (single B200)
- The mtp-nvfp4/ draft
- Deployment
- Requirements
- Single B200 / GB200 (validated)
- 2× DGX Spark (not validated)
- Request format
- Evaluation protocol
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Documentation belongs to its respective authors. Reported project/model license: mit. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
autotrust/GLM5.3-Flash-E256-DGX-Spark is a image-text-to-text model repository published by autotrust on Hugging Face. The source reports the vllm library.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- image-text-to-text
- Library
- vllm
- Recent downloads (30 days)
- 4,235
- Cumulative likes
- 118
- Hugging Face trending score
- 65
- Reported safetensors parameters
- 144,544,102,910
- Architecture
- Glm5NextForConditionalGeneration
- License
- mit
- Access
- Not gated by Hugging Face
- Created
- 2026-10-09T19:07:25.000Z
- Last modified
- 2026-10-11T03:49:59.000Z
Compatibility and lineage
Reported languages: en, zh
Reported base models: zai-org/GLM-5.3-Flash
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-10-18T06:42:00.150Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.