Store › model › LLM
GLM-5.2-GGUF
by unsloth · source Hugging Face · updated 2026-06-23
mit9 MB~1 GB RAMsource aliveunlabeled
See Unsloth Dynamic 2.0 GGUFs for our quantization benchmarks. You can now run GLM-5.2 in Unsloth Studio with toggles for High and Max thinking. Read our GLM-5.2 guide for analysis and instructions.…
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-5.2-GGUF and its progress lives in the Resource Center.
Source and license
- Source: https://huggingface.co/unsloth/GLM-5.2-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:
ggufglm_moe_dsaunslothtext-generationenzhendpoints_compatibleconversational
Numbers
- 324,023 downloads on Hugging Face
- 647 likes
- license mit
- 0.0 GB for UD-Q4_K_M/GLM-5.2-UD-Q4_K_M-00001-of-00011.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 20:53 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-5.2-BF16-00001-of-00033.gguf | BF16 | 42 GB |
BF16/GLM-5.2-BF16-00002-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00003-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00004-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00005-of-00033.gguf | BF16 | 44 GB |
BF16/GLM-5.2-BF16-00006-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00007-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00008-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00009-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00010-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00011-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00012-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00013-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00014-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00015-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00016-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00017-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00018-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00019-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00020-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00021-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00022-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00023-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00024-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00025-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00026-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00027-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00028-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00029-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00030-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00031-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00032-of-00033.gguf | BF16 | 43 GB |
BF16/GLM-5.2-BF16-00033-of-00033.gguf | BF16 | 31 GB |
Q8_0/GLM-5.2-Q8_0-00001-of-00017.gguf | Q8_0 | 45 GB |
Q8_0/GLM-5.2-Q8_0-00002-of-00017.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.2-Q8_0-00003-of-00017.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.2-Q8_0-00004-of-00017.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.2-Q8_0-00005-of-00017.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.2-Q8_0-00006-of-00017.gguf | Q8_0 | 45 GB |
Q8_0/GLM-5.2-Q8_0-00007-of-00017.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.2-Q8_0-00008-of-00017.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.2-Q8_0-00009-of-00017.gguf | Q8_0 | 45 GB |
Q8_0/GLM-5.2-Q8_0-00010-of-00017.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.2-Q8_0-00011-of-00017.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.2-Q8_0-00012-of-00017.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.2-Q8_0-00013-of-00017.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.2-Q8_0-00014-of-00017.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.2-Q8_0-00015-of-00017.gguf | Q8_0 | 45 GB |
Q8_0/GLM-5.2-Q8_0-00016-of-00017.gguf | Q8_0 | 46 GB |
Q8_0/GLM-5.2-Q8_0-00017-of-00017.gguf | Q8_0 | 16 GB |
UD-IQ1_M/GLM-5.2-UD-IQ1_M-00001-of-00006.gguf | IQ1_M | 9 MB |
UD-IQ1_M/GLM-5.2-UD-IQ1_M-00002-of-00006.gguf | IQ1_M | 46 GB |
UD-IQ1_M/GLM-5.2-UD-IQ1_M-00003-of-00006.gguf | IQ1_M | 46 GB |
UD-IQ1_M/GLM-5.2-UD-IQ1_M-00004-of-00006.gguf | IQ1_M | 46 GB |
UD-IQ1_M/GLM-5.2-UD-IQ1_M-00005-of-00006.gguf | IQ1_M | 47 GB |
UD-IQ1_M/GLM-5.2-UD-IQ1_M-00006-of-00006.gguf | IQ1_M | 29 GB |
UD-IQ1_S/GLM-5.2-UD-IQ1_S-00001-of-00006.gguf | IQ1_S | 9 MB |
UD-IQ1_S/GLM-5.2-UD-IQ1_S-00002-of-00006.gguf | IQ1_S | 46 GB |
UD-IQ1_S/GLM-5.2-UD-IQ1_S-00003-of-00006.gguf | IQ1_S | 46 GB |
UD-IQ1_S/GLM-5.2-UD-IQ1_S-00004-of-00006.gguf | IQ1_S | 46 GB |
From the source README
Read our How to Run GLM-5.2 Guide!
See Unsloth Dynamic 2.0 GGUFs for our quantization benchmarks.
You can now run GLM-5.2 in Unsloth Studio with toggles for High and Max thinking.
Read our GLM-5.2 guide for analysis and instructions.
See below for example of 1-bit UD-IQ1_M GGUF running in Unsloth:
GLM-5.2
👋 Join our WeChat or Discord community.
📖 Check out the GLM-5.2 blog and GLM-5 Technical report.
📍 Use GLM-5.2 API services on Z.ai API Platform.
🔜 Try GLM-5.2 here.
[Paper]
[GitHub]
Introduction
We're introducing GLM-5.2, our latest flagship model for long-horizon tasks. It marks a substantial leap in long-horizon task capability over its predecessor GLM-5.1 and, for the first time, delivers that capability on a solid 1M-token context. GLM-5.2's new capabilities include:
- Solid 1M Context: A solid 1M-token context that stably sustains long-horizon work
- Advanced Coding with Flexible Effort: Stronger coding capabilities with multiple thinking effort levels to balance performance and latency
- Improved Architecture: We propose IndexShare, which reuses the same indexer across every four sparse attention layers, reducing per-token FLOPs by 2.9× at a 1M context length. We also improve GLM-5.2’s MTP layer for speculative decoding, increasing the acceptance length by up to 20%
- Pure Open: An MIT open-source license — no regional limits, technical access without borders
Benchmark
|Benchmark|GLM-5.2|GLM-5.1|Qwen3.7-Max|MiniMax M3|DeepSeek-V4-Pro|Claude Opus 4.8|GPT-5.5|Gemini 3.1 Pro|
|:---|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
|Reasoning|||||||||||
|HLE|40.5|31|41.4|37|37.7|49.8*|41.4*|45|
|HLE (w/ Tools)|54.7|52.3|53.5|-|48.2|57.9*|52.2*|51.4*|
|CritPt…
Source: https://huggingface.co/unsloth/GLM-5.2-GGUF
Card id model:hf:unsloth/GLM-5.2-GGUF · collected 2026-10-02 20:53 UTC · JSON