---
title: GLM-5.3-SF-WInt4AInt8-ASC
canonical_url: "https://www.modelscope.cn/models/siliconflow/GLM-5.3-SF-WInt4AInt8-ASC"
md_url: "https://www.modelscope.cn/models/siliconflow/GLM-5.3-SF-WInt4AInt8-ASC.md"
repository: siliconflow/GLM-5.3-SF-WInt4AInt8-ASC
chinese_name: "GLM-5.3 W4A8 Ascend"
last_updated: 2026-09-08
license: other
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - glm_moe_dsa
architectures:
  - GlmMoeDsaForCausalLM
parameters: 395.1B
tensor_type:
  - I8
  - BF16
  - F32
library_name:
  - safetensors
language:
  - en
  - zh
downloads: 2606
stars: 2
---

# GLM-5.3-SF-WInt4AInt8-ASC

> GLM-5.3-SF-WInt4AInt8-ASC - siliconflow 在 ModelScope 开源的模型。GLM-5.3-SF-WInt4AInt8-ASC

siliconflow/GLM-5.3-SF-WInt4AInt8-ASC 是 ModelScope 魔搭社区上的 395.1B 参数text-generation模型，采用 other 许可。

- **Repository**: siliconflow/GLM-5.3-SF-WInt4AInt8-ASC
- **License**: other
- **Tasks**: text-generation
- **Parameters**: 395.1B
- **Downloads**: 2606
- **Stars**: 2
- **Last updated**: 2026-09-08

Source: https://www.modelscope.cn/models/siliconflow/GLM-5.3-SF-WInt4AInt8-ASC

---

# GLM-5.3-SF-WInt4AInt8-ASC

SiliconFlow / crossingbits quantization of [ZhipuAI/GLM-5.3](https://www.modelscope.cn/models/ZhipuAI/GLM-5.3) (`zai-org/GLM-5.3`).

This is the Ascend W4A8 layout (`act_group_size=128`, `backend_layout_mode=layout_only`). Source W4 `qweight` tensors are not stored in the main checkpoint; routed-expert weights live in `backend_layouts-ascend_w4a8-*.safetensors`.

## Quantization

| | |
|---|---|
| Tool | crossingbits |
| Recipe | `weight_clip` (channel-wise, `group_size=-1`, `symm_q=true`) |
| Activations | INT8, G128 |
| Save | `save_backend_layouts=[ascend_w4a8]`, `layout_only` |
| MTP | yes (`use_mtp=true`) |

### Packing

| Module | dtype |
|---|---|
| routed `mlp.experts.*.{gate,up,down}_proj` | W4A8 (Ascend sidecar) |
| L0–L2 dense MLP, shared experts, attn `q_b` / `o_proj`, indexer `wq_b` / `wk`, MTP `eh_proj` | W8A8 |
| router `mlp.gate`, MLA `q_a` / `kv_a` / `kv_b` / `k_b` / `v_b`, `indexer.weights_proj`, `shared_head`, embed, `lm_head` | BF16 |

## Base model

GLM-5.3 uses the same base as GLM-5.2; the gains are from post-training. See the [GLM-5.3 blog](https://z.ai/blog/glm-5.3) and [ZhipuAI/GLM-5.3](https://www.modelscope.cn/models/ZhipuAI/GLM-5.3).

## License

Same as the base checkpoint (`LICENSE` in this repo). Not an MIT/Apache relicense.