Ex0bit PRO

Ex0bit

32 3 18

r-e1's profile picture

Relativ3pa1n's profile picture

samiulazam's profile picture

https://ko-fi.com/ex0bit

Ex0byt
eelbaz

AI & ML interests

None yet

Recent Activity

liked a dataset 8 minutes ago

paperinstruments/diligence-bench

updated a Space 2 days ago

Ex0bit/mLattice

published a Space 2 days ago

Ex0bit/mLattice

View all activity

Organizations

Ex0bit 's collections 12

Qwen3.6 PRISM

PRISM Qwen3.6 for Mac (MLX/oMLX) and NVIDIA Blackwell NVFP4. One-click Mac setup + GPU vLLM.

Ex0bit/Qwen3.6-35B-A3B-PRISM-MLX-NVFP4

Image-Text-to-Text • 10B • Updated Jul 10 • 194 • 4
Ex0bit/Qwen3.6-35B-A3B-PRISM-NVFP4

Text Generation • Updated May 3 • 28.4k • 22

Qwen 3.5 PRISM

PRISM abliterated variants of Qwen 3.5 122B-A10B — multimodal MoE with hybrid attention. PRO, LITE, and GGUF formats.

Ex0bit/Qwen3.5-122B-A10B-PRISM-PRO-GGUF

Image-Text-to-Text • 122B • Updated Mar 2 • 176 • 27
Ex0bit/Qwen3.5-122B-A10B-PRISM-LITE-GGUF

Image-Text-to-Text • 122B • Updated Feb 28 • 95 • 16
Ex0bit/Qwen3.5-PRISM-Dynamic-Quant-GGUF

Text Generation • 0.8B • Updated Mar 3 • 466 • 11

MiniMax PRISM

PRISM abliterated MiniMax M2.1 and M2.5 MoE — reasoning, coding, agentic. GGUF and safetensors.

Ex0bit/MiniMax-M2.5-PRISM-PRO

Text Generation • 229B • Updated Feb 15 • 18
Ex0bit/MiniMax-M2.5-PRISM-LITE

Text Generation • 229B • Updated Feb 14 • 131 • 3
Ex0bit/MiniMax-M2.5-PRISM-PRO-Tensors

229B • Updated Feb 18 • 1
Ex0bit/MiniMax-M2.1-PRISM

Text Generation • 229B • Updated Jan 29 • 170 • 109

Step 3.5 PRISM

PRISM abliterated StepFun Step 3.5 Flash MoE — safetensors, GGUF, NVFP4.

Ex0bit/Step-3.5-Flash-PRISM-PRO

Text Generation • 197B • Updated Feb 9 • 4
Ex0bit/Step-3.5-Flash-PRISM

Text Generation • 197B • Updated Feb 12 • 622 • 19
Ex0bit/Step-3.5-Flash-PRISM-PRO-GGUF

Text Generation • 197B • Updated Feb 9 • 4 • 5
Ex0bit/Step-3.5-Flash-PRISM-PRO-NVFP4

111B • Updated Feb 20 • 1

OLMo 3

Allen AI OLMo 3 — abliterated 7B/32B-Think and NVFP4 quantized with 1M context.

Ex0bit/Elbaz-OLMo-3-32B-Think-Abliterated

Text Generation • 32B • Updated Nov 30, 2025 • 308 • 3
Ex0bit/Elbaz-Olmo-3-7B-Instruct-abliterated

Text Generation • 7B • Updated Nov 30, 2025 • 630 • 4
Ex0bit/OLMo-3-7B-Instruct-NVFP4-1M

Text Generation • 4B • Updated Nov 22, 2025 • 47 • 2

Papers

Research papers and experimental models.

Ex0bit/jit-lora

Updated Mar 6 • 24 • 2
Ex0bit/hrm-demo

Text Generation • Updated Sep 16, 2025 • 15 • 4

Gemma4 PRISM

Ex0bit/Gemma4-PRISM-PRO-DQ

Image-Text-to-Text • 31B • Updated Apr 8 • 3 • 32
Ex0bit/Gemma4-26B-A4B-PRISM-PRO-DQ-GGUF

Image-Text-to-Text • 25B • Updated Apr 11 • 232 • 78
Ex0bit/MYTHOS-26B-A4B-PRISM-PRO-DQ-MLX

Image-Text-to-Text • 26B • Updated Apr 11 • 463 • 25

Qwen 3 NVFP4

NVFP4-quantized Qwen 3 VLTO models for NVIDIA Blackwell / DGX Spark.

Ex0bit/Qwen3-VLTO-32B-Instruct-NVFP4

Text Generation • 17B • Updated Nov 3, 2025 • 67 • 1
Ex0bit/Qwen3-VLTO-32B-Instruct-NVFP4-256K

Text Generation • 17B • Updated Nov 4, 2025 • 31 • 1

GLM PRISM

PRISM abliterated GLM models — GLM-4.6V vision, GLM-4.7, GLM-4.7-Flash MoE.

Ex0bit/GLM-4.7-PRISM

Text Generation • 353B • Updated Jan 5 • 30
Ex0bit/GLM-4.7-Flash-PRISM

Text Generation • 30B • Updated Jan 30 • 467 • 24
Ex0bit/Elbaz-GLM-4.6V-Flash-PRISM

Image-Text-to-Text • 9B • Updated Dec 20, 2025 • 181 • 6

Kimi K2.5 PRISM

PRISM abliterated Moonshot Kimi K2.5 — multimodal MoE with REAP expert pruning.

Ex0bit/Kimi-K2.5-PRISM

Image-Text-to-Text • 171B • Updated Feb 28 • 7
Ex0bit/Kimi-K2.5-PRISM-REAP-530B-A32B

Text Generation • 91B • Updated Feb 22 • 105 • 21

NVIDIA Nemotron PRISM

PRISM abliterated NVIDIA Nemotron 3 Nano 30B-A3B — Mamba-2 hybrid MoE.

Ex0bit/Elbaz-NVIDIA-Nemotron-3-Nano-30B-A3B-PRISM

Text Generation • 32B • Updated Jan 23 • 480 • 30
Ex0bit/Elbaz-NVIDIA-Nemotron-3-Nano-30B-A3B-PRISM-NVFP4

Text Generation • 16B • Updated Dec 18, 2025 • 57 • 6

Research Models

Experimental and research model variants.

Ex0bit/lfm-Nanotron

Text Generation • Updated Jan 24 • 42 • 3
Ex0bit/Elbaz-Prime-Intellect-3_Prism_Abliterated

Text Generation • 107B • Updated Dec 3, 2025 • 93 • 5

Qwen3.6 PRISM

PRISM Qwen3.6 for Mac (MLX/oMLX) and NVIDIA Blackwell NVFP4. One-click Mac setup + GPU vLLM.

Ex0bit/Qwen3.6-35B-A3B-PRISM-MLX-NVFP4

Image-Text-to-Text • 10B • Updated Jul 10 • 194 • 4
Ex0bit/Qwen3.6-35B-A3B-PRISM-NVFP4

Text Generation • Updated May 3 • 28.4k • 22

Gemma4 PRISM

Ex0bit/Gemma4-PRISM-PRO-DQ

Image-Text-to-Text • 31B • Updated Apr 8 • 3 • 32
Ex0bit/Gemma4-26B-A4B-PRISM-PRO-DQ-GGUF

Image-Text-to-Text • 25B • Updated Apr 11 • 232 • 78
Ex0bit/MYTHOS-26B-A4B-PRISM-PRO-DQ-MLX

Image-Text-to-Text • 26B • Updated Apr 11 • 463 • 25

Qwen 3.5 PRISM

PRISM abliterated variants of Qwen 3.5 122B-A10B — multimodal MoE with hybrid attention. PRO, LITE, and GGUF formats.

Ex0bit/Qwen3.5-122B-A10B-PRISM-PRO-GGUF

Image-Text-to-Text • 122B • Updated Mar 2 • 176 • 27
Ex0bit/Qwen3.5-122B-A10B-PRISM-LITE-GGUF

Image-Text-to-Text • 122B • Updated Feb 28 • 95 • 16
Ex0bit/Qwen3.5-PRISM-Dynamic-Quant-GGUF

Text Generation • 0.8B • Updated Mar 3 • 466 • 11

Qwen 3 NVFP4

NVFP4-quantized Qwen 3 VLTO models for NVIDIA Blackwell / DGX Spark.

Ex0bit/Qwen3-VLTO-32B-Instruct-NVFP4

Text Generation • 17B • Updated Nov 3, 2025 • 67 • 1
Ex0bit/Qwen3-VLTO-32B-Instruct-NVFP4-256K

Text Generation • 17B • Updated Nov 4, 2025 • 31 • 1

MiniMax PRISM

PRISM abliterated MiniMax M2.1 and M2.5 MoE — reasoning, coding, agentic. GGUF and safetensors.

Ex0bit/MiniMax-M2.5-PRISM-PRO

Text Generation • 229B • Updated Feb 15 • 18
Ex0bit/MiniMax-M2.5-PRISM-LITE

Text Generation • 229B • Updated Feb 14 • 131 • 3
Ex0bit/MiniMax-M2.5-PRISM-PRO-Tensors

229B • Updated Feb 18 • 1
Ex0bit/MiniMax-M2.1-PRISM

Text Generation • 229B • Updated Jan 29 • 170 • 109

GLM PRISM

PRISM abliterated GLM models — GLM-4.6V vision, GLM-4.7, GLM-4.7-Flash MoE.

Ex0bit/GLM-4.7-PRISM

Text Generation • 353B • Updated Jan 5 • 30
Ex0bit/GLM-4.7-Flash-PRISM

Text Generation • 30B • Updated Jan 30 • 467 • 24
Ex0bit/Elbaz-GLM-4.6V-Flash-PRISM

Image-Text-to-Text • 9B • Updated Dec 20, 2025 • 181 • 6

Step 3.5 PRISM

PRISM abliterated StepFun Step 3.5 Flash MoE — safetensors, GGUF, NVFP4.

Ex0bit/Step-3.5-Flash-PRISM-PRO

Text Generation • 197B • Updated Feb 9 • 4
Ex0bit/Step-3.5-Flash-PRISM

Text Generation • 197B • Updated Feb 12 • 622 • 19
Ex0bit/Step-3.5-Flash-PRISM-PRO-GGUF

Text Generation • 197B • Updated Feb 9 • 4 • 5
Ex0bit/Step-3.5-Flash-PRISM-PRO-NVFP4

111B • Updated Feb 20 • 1

Kimi K2.5 PRISM

PRISM abliterated Moonshot Kimi K2.5 — multimodal MoE with REAP expert pruning.

Ex0bit/Kimi-K2.5-PRISM

Image-Text-to-Text • 171B • Updated Feb 28 • 7
Ex0bit/Kimi-K2.5-PRISM-REAP-530B-A32B

Text Generation • 91B • Updated Feb 22 • 105 • 21

OLMo 3

Allen AI OLMo 3 — abliterated 7B/32B-Think and NVFP4 quantized with 1M context.

Ex0bit/Elbaz-OLMo-3-32B-Think-Abliterated

Text Generation • 32B • Updated Nov 30, 2025 • 308 • 3
Ex0bit/Elbaz-Olmo-3-7B-Instruct-abliterated

Text Generation • 7B • Updated Nov 30, 2025 • 630 • 4
Ex0bit/OLMo-3-7B-Instruct-NVFP4-1M

Text Generation • 4B • Updated Nov 22, 2025 • 47 • 2

NVIDIA Nemotron PRISM

PRISM abliterated NVIDIA Nemotron 3 Nano 30B-A3B — Mamba-2 hybrid MoE.

Ex0bit/Elbaz-NVIDIA-Nemotron-3-Nano-30B-A3B-PRISM

Text Generation • 32B • Updated Jan 23 • 480 • 30
Ex0bit/Elbaz-NVIDIA-Nemotron-3-Nano-30B-A3B-PRISM-NVFP4

Text Generation • 16B • Updated Dec 18, 2025 • 57 • 6

Papers

Research papers and experimental models.

Ex0bit/jit-lora

Updated Mar 6 • 24 • 2
Ex0bit/hrm-demo

Text Generation • Updated Sep 16, 2025 • 15 • 4

Research Models

Experimental and research model variants.

Ex0bit/lfm-Nanotron

Text Generation • Updated Jan 24 • 42 • 3
Ex0bit/Elbaz-Prime-Intellect-3_Prism_Abliterated

Text Generation • 107B • Updated Dec 3, 2025 • 93 • 5