license: mit
base_model:
- dealignai/GLM-5.3-Flash-UNCENSORED-FP8
- zai-org/GLM-5.3-Flash
tags: - gguf
- glm
- moe
- uncensored
- ik_llama.cpp
- g53
GLM-5.3-Flash-UNCENSORED — R2-40 IQ5_KS expert GGUF (for g53)
Routed-expert weights for running GLM-5.3-Flash (320B total / 18B active MoE) on one or two RTX 3090s with the
g53 engine. g53 reads the routed experts from this file (CPU + VRAM hot cache)
and the dense backbone from the FP8 checkpoint dealignai/GLM-5.3-Flash-UNCENSORED-FP8.
- Base: zai-org/GLM-5.3-Flash (MIT), uncensored derivative by
dealignai (MIT). This is an uncensored model:
safety refusals were removed by the derivative's author. Use responsibly. - Quantization: ik_llama.cpp
llama-quantize, routed experts IQ5_KS
(last MoE layer IQ6_K), shared experts / dense FFN Q8_0, attention and indexer BF16; ~207 GB in 13 shards;
importance matrix by AesSedai. Full recipe:
--custom-q 'blk\.45\.ffn_(gate|up|down)_exps\.weight=iq6_k,ffn_(gate|up|down)_exps\.weight=iq5_ks,ffn_(gate|up|down)_shexp\.weight=q8_0,ffn_(gate|up|down)\.weight=q8_0,indexer\.(attn_k|attn_q_b|proj)\.weight|indexer_compressor_(gate|ape)\.weight=bf16,hc_(attn|ffn)_fn\.weight=f32,attn_(k_b|v_b)\.weight=bf16,attn_(q_a|q_b)\.weight=q8_0,attn_kv_a_mqa\.weight=bf16,attn_(q|k|v|output)\.weight=bf16,ssm_(f_a|f_b|g_a|g_b|beta)\.weight=bf16,nextn\.eh_proj\.weight=bf16,token_embd\.weight=q8_0,output\.weight=bf16'
- Quality: mean KL divergence vs BF16 of 0.026–0.036 on held-out text, on par with ik_llama.cpp on the same file.
- Also included:
mmproj-GLM-5.3-Flash-UNCENSORED-FP8-BF16.gguf, the vision projector (BF16, converted with
llama.cppconvert_hf_to_gguf.py --mmproj), used by g53 for image input.
Use with g53
hf download ily-lab/GLM-5.3-Flash-UNCENSORED-R2-40-GGUF --local-dir models/experts
hf download dealignai/GLM-5.3-Flash-UNCENSORED-FP8 --local-dir models/GLM-5.3-Flash-FP8
export G53_EXPERTS=models/experts/GLM-5.3-Flash-UNCENSORED-R2-40-00001-of-00013.gguf
export G53_MMPROJ=models/experts/mmproj-GLM-5.3-Flash-UNCENSORED-FP8-BF16.gguf
G53_GPU1=1 tools/g53-serve.sh # 2× RTX 3090; omit G53_GPU1 for one GPU
The expert file is also a valid ik_llama.cpp GGUF of the full model (it contains all tensors).
License
MIT, inherited from GLM-5.3-Flash (Copyright (c) 2026 Z.AI Co., Ltd) and the dealignai derivative.