Find the right release.

MoEPUBLISHEDQ4_K_M · 14.5 GB

LFM2.5-24B-A2B MoE GGUF

Certified GGUF precision quantization ladder for LiquidAI/LFM2.5-24B-A2B hybrid convolution and mixture-of-experts architecture.

HFOmega-Factory/LFM2.5-24B-A2B-GGUF

Parameters: 24.00B(2.45B act)
Q4_K_M Size: 14.5 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 72.7%

ARC-MC

89.0%+1.0

GSM8K

78.5%+0.0

MATH-500

50.5%-8.0

Q4_K_MQ8_0Q2_K

LFM Open License v1.0

DensePUBLISHEDQ4_K_M · 17.7 GB

Granite 4.2 30B GGUF

Certified GGUF precision quantization ladder for ibm-granite/granite-4.2-30b dense enterprise architecture.

HFOmega-Factory/granite-4.2-30b-GGUF

Parameters: 29.30B
Q4_K_M Size: 17.7 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 79.3%

ARC-MC

91.0%-1.0

GSM8K

92.5%-1.5

MATH-500

54.5%+0.0

Q4_K_MQ8_0Q2_K

DensePUBLISHEDQ4_K_M · 16.9 GB

Muse Glimmer 30B GGUF

Certified GGUF precision quantization ladder for meta-models/Muse-Glimmer-30B dense multimodal architecture.

HFOmega-Factory/muse-glimmer-30b-GGUF

Parameters: 29.60B
Q4_K_M Size: 16.9 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 83.7%

ARC-MC

96.0%+0.5

GSM8K

93.0%-2.5

MATH-500

62.0%-3.0

Q4_K_MQ8_0Q2_K

DensePUBLISHEDQ4_K_M · 18.7 GB

Gemma 4 31B Dense GGUF

Certified GGUF precision quantization ladder for google/gemma-4-31b dense architecture.

HFOmega-Factory/gemma-4-31b-GGUF

Parameters: 30.70B
Q4_K_M Size: 18.7 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 72.5%

ARC-MC

94.5%+9.5

GSM8K

87.5%+1.5

MATH-500

35.5%-1.0

Q4_K_MQ8_0Q2_K

MoEPUBLISHEDQ4_K_M · 16.8 GB

Gemma 4 26B-A4B MoE GGUF

Certified GGUF precision quantization ladder for google/gemma-4-26b-a4b mixture-of-experts architecture.

HFOmega-Factory/gemma-4-26b-a4b-GGUF

Parameters: 25.23B(3.80B act)
Q4_K_M Size: 16.8 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 60.5%

ARC-MC

92.5%+0.5

GSM8K

66.0%-7.0

MATH-500

23.0%-13.5

Q4_K_MQ8_0Q2_K

MoEPUBLISHEDQ4_K_M · 14.2 GB

Qwen3-30B-A3B GGUF

Certified GGUF precision quantization ladder for Qwen/Qwen3-30B-A3B pruned and healed MoE architecture.

HFOmega-Factory/Qwen3-30B-A3B-GGUF

Parameters: 23.28B(3.30B act)
Q4_K_M Size: 14.2 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 78.3%

ARC-MC

93.0%-0.5

GSM8K

94.0%+1.5

MATH-500

48.0%-1.5

Q4_K_MQ8_0Q2_K

DensePUBLISHEDQ4_K_M · 16.8 GB

Qwen3.8-27B GGUF

Certified GGUF precision quantization ladder for Qwen/Qwen3.8-27B dense vision-language model.

HFOmega-Factory/Qwen3.8-27B-GGUF

Parameters: 27.32B
Q4_K_M Size: 16.8 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 48.7%

ARC-MC

95.0%+0.5

GSM8K

33.5%-25.5

MATH-500

17.5%-5.5

Q4_K_MQ8_0Q2_K

Qwen3-Coder-Next MoE GGUF

Cross-Model Benchmark Matrix

Side-by-side precision & accuracy comparison

Deterministic N=200 holdout evaluations (seed 42) across ARC-Challenge, GSM8K, and MATH-500.

Model	Arch / Params	Q4_K_M Size	ARC-MC	GSM8K	MATH-500	Mean	License
Qwen3-Coder-Next MoE GGUF	moe · 79.60B (3.14B act)	48.49 GB	97.0%(+0.5)	95.5%	61.0%(-0.5)	84.5%	Apache 2.0
LFM2.5-24B-A2B MoE GGUF	moe · 24.00B (2.45B act)	14.53 GB	89.0%(+1.0)	78.5%	50.5%(-8.0)	72.7%	LFM Open License v1.0
Granite 4.2 30B GGUF	dense · 29.30B	17.72 GB	91.0%(-1.0)	92.5%(-1.5)	54.5%	79.3%	Apache 2.0
Muse Glimmer 30B GGUF	dense · 29.60B	16.94 GB	96.0%(+0.5)	93.0%(-2.5)	62.0%(-3.0)	83.7%	Apache 2.0
Gemma 4 31B Dense GGUF	dense · 30.70B	18.69 GB	94.5%(+9.5)	87.5%(+1.5)	35.5%(-1.0)	72.5%	Gemma License
Gemma 4 26B-A4B MoE GGUF	moe · 25.23B (3.80B act)	16.80 GB	92.5%(+0.5)	66.0%(-7.0)	23.0%(-13.5)	60.5%	Gemma License
Qwen3-30B-A3B GGUF	moe · 23.28B (3.30B act)	14.16 GB	93.0%(-0.5)	94.0%(+1.5)	48.0%(-1.5)	78.3%	Apache 2.0
Qwen3.8-27B GGUF	dense · 27.32B	16.81 GB	95.0%(+0.5)	33.5%(-25.5)	17.5%(-5.5)	48.7%	Apache 2.0

moe · 79.60B (3.14B act) · 48.49 GB

Mean 84.5%

ARC-MC

97.0%

+0.5 pp

GSM8K

95.5%

MATH-500

61.0%

-0.5 pp

LFM2.5-24B-A2B MoE GGUF

moe · 24.00B (2.45B act) · 14.53 GB

Mean 72.7%

ARC-MC

89.0%

+1.0 pp

GSM8K

78.5%

MATH-500

50.5%

-8.0 pp

Granite 4.2 30B GGUF

dense · 29.30B · 17.72 GB

Mean 79.3%

ARC-MC

91.0%

-1.0 pp

GSM8K

92.5%

-1.5 pp

MATH-500

54.5%

Muse Glimmer 30B GGUF

dense · 29.60B · 16.94 GB

Mean 83.7%

ARC-MC

96.0%

+0.5 pp

GSM8K

93.0%

-2.5 pp

MATH-500

62.0%

-3.0 pp

Gemma 4 31B Dense GGUF

dense · 30.70B · 18.69 GB

Mean 72.5%

ARC-MC

94.5%

+9.5 pp

GSM8K

87.5%

+1.5 pp

MATH-500

35.5%

-1.0 pp

Gemma 4 26B-A4B MoE GGUF

moe · 25.23B (3.80B act) · 16.80 GB

Mean 60.5%

ARC-MC

92.5%

+0.5 pp

GSM8K

66.0%

-7.0 pp

MATH-500

23.0%

-13.5 pp

Qwen3-30B-A3B GGUF

moe · 23.28B (3.30B act) · 14.16 GB

Mean 78.3%

ARC-MC

93.0%

-0.5 pp

GSM8K

94.0%

+1.5 pp

MATH-500

48.0%

-1.5 pp

Qwen3.8-27B GGUF

dense · 27.32B · 16.81 GB

Mean 48.7%

ARC-MC

95.0%

+0.5 pp

GSM8K

33.5%

-25.5 pp

MATH-500

17.5%

-5.5 pp

Catalog public_model_catalog_v1 · generated 2026-10-14T08:00:00Z · sha256 d853e89023b5b0378c96b28cf519639e2b0bb9b774d171f1fe1e0dbcc48b68f8

Checking the public catalog and exact artifact bindings…

Public model catalog

Find the right release.

Search verified public releases by model, architecture, artifact format, or license. Each detail page binds downloads and scoped measurements to exact immutable identities.

Approved catalog records only Immutable Hugging Face revisionsGGUF Quantization & Benchmark Docs →

Sort

8 models

Newest first

MoEPUBLISHEDQ4_K_M · 48.5 GB

Qwen3-Coder-Next MoE GGUF

Certified GGUF precision quantization ladder for Qwen/Qwen3-Coder-Next hybrid Gated DeltaNet and mixture-of-experts coding model.

HFOmega-Factory/Qwen3-Coder-Next-GGUF

Parameters: 79.60B(3.14B act)
Q4_K_M Size: 48.5 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 84.5%

ARC-MC

97.0%+0.5

GSM8K

95.5%+0.0

MATH-500

61.0%-0.5

Q4_K_MQ8_0Q2_K

MoEPUBLISHEDQ4_K_M · 14.5 GB

LFM2.5-24B-A2B MoE GGUF

Certified GGUF precision quantization ladder for LiquidAI/LFM2.5-24B-A2B hybrid convolution and mixture-of-experts architecture.

HFOmega-Factory/LFM2.5-24B-A2B-GGUF

Parameters: 24.00B(2.45B act)
Q4_K_M Size: 14.5 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 72.7%

ARC-MC

89.0%+1.0

GSM8K

78.5%+0.0

MATH-500

50.5%-8.0

Q4_K_MQ8_0Q2_K

LFM Open License v1.0

DensePUBLISHEDQ4_K_M · 17.7 GB

Granite 4.2 30B GGUF

Certified GGUF precision quantization ladder for ibm-granite/granite-4.2-30b dense enterprise architecture.

HFOmega-Factory/granite-4.2-30b-GGUF

Parameters: 29.30B
Q4_K_M Size: 17.7 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 79.3%

ARC-MC

91.0%-1.0

GSM8K

92.5%-1.5

MATH-500

54.5%+0.0

Q4_K_MQ8_0Q2_K

DensePUBLISHEDQ4_K_M · 16.9 GB

Muse Glimmer 30B GGUF

Certified GGUF precision quantization ladder for meta-models/Muse-Glimmer-30B dense multimodal architecture.

HFOmega-Factory/muse-glimmer-30b-GGUF

Parameters: 29.60B
Q4_K_M Size: 16.9 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 83.7%

ARC-MC

96.0%+0.5

GSM8K

93.0%-2.5

MATH-500

62.0%-3.0

Q4_K_MQ8_0Q2_K

DensePUBLISHEDQ4_K_M · 18.7 GB

Gemma 4 31B Dense GGUF

Certified GGUF precision quantization ladder for google/gemma-4-31b dense architecture.

HFOmega-Factory/gemma-4-31b-GGUF

Parameters: 30.70B
Q4_K_M Size: 18.7 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 72.5%

ARC-MC

94.5%+9.5

GSM8K

87.5%+1.5

MATH-500

35.5%-1.0

Q4_K_MQ8_0Q2_K

MoEPUBLISHEDQ4_K_M · 16.8 GB

Gemma 4 26B-A4B MoE GGUF

Certified GGUF precision quantization ladder for google/gemma-4-26b-a4b mixture-of-experts architecture.

HFOmega-Factory/gemma-4-26b-a4b-GGUF

Parameters: 25.23B(3.80B act)
Q4_K_M Size: 16.8 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 60.5%

ARC-MC

92.5%+0.5

GSM8K

66.0%-7.0

MATH-500

23.0%-13.5

Q4_K_MQ8_0Q2_K

MoEPUBLISHEDQ4_K_M · 14.2 GB

Qwen3-30B-A3B GGUF

Certified GGUF precision quantization ladder for Qwen/Qwen3-30B-A3B pruned and healed MoE architecture.

HFOmega-Factory/Qwen3-30B-A3B-GGUF

Parameters: 23.28B(3.30B act)
Q4_K_M Size: 14.2 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 78.3%

ARC-MC

93.0%-0.5

GSM8K

94.0%+1.5

MATH-500

48.0%-1.5

Q4_K_MQ8_0Q2_K

DensePUBLISHEDQ4_K_M · 16.8 GB

Qwen3.8-27B GGUF

Certified GGUF precision quantization ladder for Qwen/Qwen3.8-27B dense vision-language model.

HFOmega-Factory/Qwen3.8-27B-GGUF

Parameters: 27.32B
Q4_K_M Size: 16.8 GB
Artifacts: 3 GGUFs

Q4_K_M Holdout Benchmarks (N=200)Avg 48.7%

ARC-MC

95.0%+0.5

GSM8K

33.5%-25.5

MATH-500

17.5%-5.5

Q4_K_MQ8_0Q2_K