Store › model › LLM
GLM-5.3-Flash-GGUF
by unsloth · source Hugging Face · updated 2026-09-06
mit9 MB~1 GB RAMsource aliveunlabeled
Unsloth Dynamic 3.0 achieves superior accuracy & outperforms other leading quants. To run, please use our llama.cpp PR or use the Unsloth Desktop app. You can now run GLM-5.3-Flash in our Unsloth Des…
Add to LogiShell Open in the web IDE
The button opens LogiShell with this card; nothing installs from a link by itself. Inside the app the install goes through lsh models install hf:unsloth/GLM-5.3-Flash-GGUF and its progress lives in the Resource Center.
Source and license
- Source: https://huggingface.co/unsloth/GLM-5.3-Flash-GGUF
- License: mit
- Requirements: about 1 GB of RAM, 9 MB on disk (estimate: suggested file size × 1.15 + 0.5 GB; a real measurement comes with lsh models). Runs with llama.cpp, ollama.
- Tags:
ggufunslothglm5_nexttext-generationenzhendpoints_compatibleconversational
Numbers
- 1,012,847 downloads on Hugging Face
- 443 likes
- license mit
- 0.0 GB for Q8_0/GLM-5.3-Flash-Q8_0-00001-of-00008.gguf
- 130,156 stars on ggml-org/llama.cpp
- 2,528 open issues and PRs
- last release v0.5.0 on 2026-09-23
- 327,838 npm downloads a week for node-llama-cpp
- latest node-llama-cpp@3.22.1
Numbers as of 2026-10-02 20:52 UTC, from the source API.
Summary
Summary not ready yet: the numbers are here, the text is not. It is written by the collector through the LogiShell model facade when a provider key is present.
Reviews
No reviews yet. Reviews are written inside LogiShell: open this card in the app.
Files
| file | quant | size |
|---|---|---|
BF16/GLM-5.3-Flash-BF16-00001-of-00014.gguf | BF16 | 9 MB |
BF16/GLM-5.3-Flash-BF16-00002-of-00014.gguf | BF16 | 45 GB |
BF16/GLM-5.3-Flash-BF16-00003-of-00014.gguf | BF16 | 46 GB |
BF16/GLM-5.3-Flash-BF16-00004-of-00014.gguf | BF16 | 46 GB |
BF16/GLM-5.3-Flash-BF16-00005-of-00014.gguf | BF16 | 47 GB |
BF16/GLM-5.3-Flash-BF16-00006-of-00014.gguf | BF16 | 46 GB |
BF16/GLM-5.3-Flash-BF16-00007-of-00014.gguf | BF16 | 46 GB |
BF16/GLM-5.3-Flash-BF16-00008-of-00014.gguf | BF16 | 46 GB |
BF16/GLM-5.3-Flash-BF16-00009-of-00014.gguf | BF16 | 46 GB |
BF16/GLM-5.3-Flash-BF16-00010-of-00014.gguf | BF16 | 46 GB |
BF16/GLM-5.3-Flash-BF16-00011-of-00014.gguf | BF16 | 46 GB |
BF16/GLM-5.3-Flash-BF16-00012-of-00014.gguf | BF16 | 46 GB |
BF16/GLM-5.3-Flash-BF16-00013-of-00014.gguf | BF16 | 46 GB |
BF16/GLM-5.3-Flash-BF16-00014-of-00014.gguf | BF16 | 46 GB |
Q8_0/GLM-5.3-Flash-Q8_0-00001-of-00008.gguf | Q8_0 | 9 MB |
Q8_0/GLM-5.3-Flash-Q8_0-00002-of-00008.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.3-Flash-Q8_0-00003-of-00008.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.3-Flash-Q8_0-00004-of-00008.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.3-Flash-Q8_0-00005-of-00008.gguf | Q8_0 | 47 GB |
Q8_0/GLM-5.3-Flash-Q8_0-00006-of-00008.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.3-Flash-Q8_0-00007-of-00008.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.3-Flash-Q8_0-00008-of-00008.gguf | Q8_0 | 39 GB |
UD-IQ1_M/GLM-5.3-Flash-UD-IQ1_M-00001-of-00003.gguf | IQ1_M | 9 MB |
UD-IQ1_M/GLM-5.3-Flash-UD-IQ1_M-00002-of-00003.gguf | IQ1_M | 47 GB |
UD-IQ1_M/GLM-5.3-Flash-UD-IQ1_M-00003-of-00003.gguf | IQ1_M | 44 GB |
UD-IQ1_S/GLM-5.3-Flash-UD-IQ1_S-00001-of-00003.gguf | IQ1_S | 9 MB |
UD-IQ1_S/GLM-5.3-Flash-UD-IQ1_S-00002-of-00003.gguf | IQ1_S | 46 GB |
UD-IQ1_S/GLM-5.3-Flash-UD-IQ1_S-00003-of-00003.gguf | IQ1_S | 40 GB |
UD-IQ2_XXS/GLM-5.3-Flash-UD-IQ2_XXS-00001-of-00004.gguf | IQ2_XXS | 9 MB |
UD-IQ2_XXS/GLM-5.3-Flash-UD-IQ2_XXS-00002-of-00004.gguf | IQ2_XXS | 46 GB |
UD-IQ2_XXS/GLM-5.3-Flash-UD-IQ2_XXS-00003-of-00004.gguf | IQ2_XXS | 46 GB |
UD-IQ2_XXS/GLM-5.3-Flash-UD-IQ2_XXS-00004-of-00004.gguf | IQ2_XXS | 2.5 GB |
UD-IQ3_XXS/GLM-5.3-Flash-UD-IQ3_XXS-00001-of-00004.gguf | IQ3_XXS | 9 MB |
UD-IQ3_XXS/GLM-5.3-Flash-UD-IQ3_XXS-00002-of-00004.gguf | IQ3_XXS | 46 GB |
UD-IQ3_XXS/GLM-5.3-Flash-UD-IQ3_XXS-00003-of-00004.gguf | IQ3_XXS | 46 GB |
UD-IQ3_XXS/GLM-5.3-Flash-UD-IQ3_XXS-00004-of-00004.gguf | IQ3_XXS | 21 GB |
UD-IQ4_XS/GLM-5.3-Flash-UD-IQ4_XS-00001-of-00005.gguf | IQ4_XS | 9 MB |
UD-IQ4_XS/GLM-5.3-Flash-UD-IQ4_XS-00002-of-00005.gguf | IQ4_XS | 47 GB |
UD-IQ4_XS/GLM-5.3-Flash-UD-IQ4_XS-00003-of-00005.gguf | IQ4_XS | 46 GB |
UD-IQ4_XS/GLM-5.3-Flash-UD-IQ4_XS-00004-of-00005.gguf | IQ4_XS | 46 GB |
UD-IQ4_XS/GLM-5.3-Flash-UD-IQ4_XS-00005-of-00005.gguf | IQ4_XS | 7.2 GB |
UD-Q2_K_XL/GLM-5.3-Flash-UD-Q2_K_XL-00001-of-00004.gguf | Q2_K_XL | 9 MB |
UD-Q2_K_XL/GLM-5.3-Flash-UD-Q2_K_XL-00002-of-00004.gguf | Q2_K_XL | 46 GB |
UD-Q2_K_XL/GLM-5.3-Flash-UD-Q2_K_XL-00003-of-00004.gguf | Q2_K_XL | 47 GB |
UD-Q2_K_XL/GLM-5.3-Flash-UD-Q2_K_XL-00004-of-00004.gguf | Q2_K_XL | 8.8 GB |
UD-Q3_K_XL/GLM-5.3-Flash-UD-Q3_K_XL-00001-of-00004.gguf | Q3_K_XL | 9 MB |
UD-Q3_K_XL/GLM-5.3-Flash-UD-Q3_K_XL-00002-of-00004.gguf | Q3_K_XL | 46 GB |
UD-Q3_K_XL/GLM-5.3-Flash-UD-Q3_K_XL-00003-of-00004.gguf | Q3_K_XL | 46 GB |
UD-Q3_K_XL/GLM-5.3-Flash-UD-Q3_K_XL-00004-of-00004.gguf | Q3_K_XL | 45 GB |
UD-Q4_K_XL/GLM-5.3-Flash-UD-Q4_K_XL-00001-of-00006.gguf | Q4_K_XL | 9 MB |
UD-Q4_K_XL/GLM-5.3-Flash-UD-Q4_K_XL-00002-of-00006.gguf | Q4_K_XL | 46 GB |
UD-Q4_K_XL/GLM-5.3-Flash-UD-Q4_K_XL-00003-of-00006.gguf | Q4_K_XL | 47 GB |
UD-Q4_K_XL/GLM-5.3-Flash-UD-Q4_K_XL-00004-of-00006.gguf | Q4_K_XL | 45 GB |
UD-Q4_K_XL/GLM-5.3-Flash-UD-Q4_K_XL-00005-of-00006.gguf | Q4_K_XL | 46 GB |
UD-Q4_K_XL/GLM-5.3-Flash-UD-Q4_K_XL-00006-of-00006.gguf | Q4_K_XL | 2.6 GB |
UD-Q5_K_XL/GLM-5.3-Flash-UD-Q5_K_XL-00001-of-00006.gguf | Q5_K_XL | 9 MB |
UD-Q5_K_XL/GLM-5.3-Flash-UD-Q5_K_XL-00002-of-00006.gguf | Q5_K_XL | 45 GB |
UD-Q5_K_XL/GLM-5.3-Flash-UD-Q5_K_XL-00003-of-00006.gguf | Q5_K_XL | 45 GB |
UD-Q5_K_XL/GLM-5.3-Flash-UD-Q5_K_XL-00004-of-00006.gguf | Q5_K_XL | 46 GB |
UD-Q5_K_XL/GLM-5.3-Flash-UD-Q5_K_XL-00005-of-00006.gguf | Q5_K_XL | 46 GB |
From the source README
Read our How to Run GLM-5.3-Flash Guide!
Unsloth Dynamic 3.0 achieves superior accuracy & outperforms other leading quants.
To run, please use our llama.cpp PR or use the Unsloth Desktop app.
You can now run GLM-5.3-Flash in our Unsloth Desktop UI.
See below for 1-bit GLM-5.3-Flash (Low) run inside of Unsloth Desktop:
GLM-5.3-Flash
👋 Join our WeChat or Discord community.
📖 Check out the GLM-5.3-Flash blog and GLM-5 Technical report.
📍 Use GLM-5.3-Flash API services on Z.ai API Platform.
Introduction
We introduce GLM-5.3-Flash, the first natively multimodal model in the GLM-5 series. With 320B total parameters and just 18B active parameters, it outperforms GLM-5.2 across benchmarks and real-world workloads at one-tenth the price, while approaching Claude Opus 4.8 on coding and agentic benchmarks.
GLM-5.3-Flash starts from a newly trained base model, with its architecture and training recipe redesigned around capability and efficiency. For the first time in the GLM series, we introduce a hybrid architecture combining sparse and linear attention, sharply reducing long-context serving costs while preserving precise long-context capabilities. The model also adopts Manifold-Constrained Hyper-Connections (mHC) to further improve scaling efficiency. Together with our latest 30T-token multimodal pre-training corpus, these changes enable GLM-5.3-Flash to deliver more intelligence with less compute.
Footnotes
- HLE w/ tools (full set): We use sampling parameters of `temperature=1.0` and `top_p=0.95` for evaluation, with a maximum generation length of `163,840` tokens. The evaluation is conducted with a maximum context length of `300,000` tokens, using a context management strategy. We use GPT-5.6-luna (medium) as the judge model.
- **NL2…
Source: https://huggingface.co/unsloth/GLM-5.3-Flash-GGUF
Card id model:hf:unsloth/GLM-5.3-Flash-GGUF · collected 2026-10-02 20:52 UTC · JSON