orcarouter_GLM-5.3-Flash-Uncensored-GGUF
GGUF quantizations of orcarouter/GLM-5.3-Flash-Uncensored-FP8, built with a custom importance matrix and measured rather than assumed.
Params
321.3 B
Context
—
Downloads 30d
492 K
Likes
31
Download history
daily snapshots · 18 days492 K351 K
Oct 7Oct 13Oct 19Oct 24
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| mmproj-GLM-5.3-Flash-Uncensored-F16.gguf | GGUF | 1.1 GB | 49.9 GB | ❌ Won’t fit |
| GLM-5.3-Flash-Uncensored-IQ2_S-00001-of-00003.gguf | IQ2_S | 36.0 GB | 88.3 GB | ❌ Won’t fit |
| GLM-5.3-Flash-Uncensored-IQ2_XXS-00001-of-00003.gguf | IQ2_XXS | 36.0 GB | 88.3 GB | ❌ Won’t fit |
| GLM-5.3-Flash-Uncensored-IQ2_M-00002-of-00003.gguf | IQ2_M | 41.7 GB | 94.5 GB | ❌ Won’t fit |
| GLM-5.3-Flash-Uncensored-IQ3_XXS-00001-of-00003.gguf | IQ3_XXS | 45.7 GB | 98.9 GB | ❌ Won’t fit |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Run it
copy-paste, exact tags checked against the Hub$ ollama run orcarouter-glm-5-3-flash-uncensored-gguf # pin the quantization explicitly $ ollama run orcarouter-glm-5-3-flash-uncensored-gguf-iq2_m
$ huggingface-cli download BoldingBuilds/orcarouter_GLM-5.3-Flash-Uncensored-GGUF-GGUF \
GLM-5.3-Flash-Uncensored-IQ2_M-00002-of-00003.gguf --local-dir .
$ llama-cli -m GLM-5.3-Flash-Uncensored-IQ2_M-00002-of-00003.gguf \
-c 8192 -ngl 99 -t 8 --color
$ curl -s https://aimodelscomparison.com/api/v1/models/orcarouter-glm-5-3-flash-uncensored-gguf
{
"hf_id": "BoldingBuilds/orcarouter_GLM-5.3-Flash-Uncensored-GGUF",
"params_b": 321.32,
"context_length": null,
"license": { "id": "mit", "commercial": "yes" },
"downloads_30d": 491636,
"vram_estimates": [
{ "quant": "IQ2_M", "gb": 94.5 },
{ "quant": "IQ2_S", "gb": 88.3 }
],
"updated_at": "2026-10-09T01:00:11Z"
}
Specifications
- Parameters
- 321.3 B
- Licence
- mit
- First seen on the Hub
- 2026-08-31
- Base model
- GLM-5.3-Flash-Uncensored-FP8
- Training datasets
- undisclosed
- Added to our catalog
- 2026-10-07
Family
Base model and the most-downloaded derivatives in the catalog.
Compare with any text-generation model