Recursive's Flint for TandemLLM on one DGX Spark: our EXL3 packs at 4.0 (the default) and 6.0 bits a weight, and Recursive's BF16 checkpoint.
0xBakeer PRO
0xBakeer
AI & ML interests
None yet
Recent Activity
liked a model 5 days ago
ricursiveAI/Ripple-65B-Preview-NVFP4 liked a model 5 days ago
moonshotai/Kimi-K3.1-Pro updated a model 7 days ago
0xBakeer/TandemLLM-Qwen3.8-Flash-Next-EXL3-3.05-correctorOrganizations
EXL3 packs of NVIDIA Nemotron 3.5 Super VL 120B-A12B at 4.0 and 2.0 bpw, for TandemLLM 0.11.0 on DGX Spark. OpenMDW-1.1.
-
0xBakeer/TandemLLM-Nemotron-3.5-Super-VL-120B-A12B-EXL3-4.0
Text Generation • 32B • Updated • 258 • 1 -
0xBakeer/TandemLLM-Nemotron-3.5-Super-VL-120B-A12B-EXL3-2.0
Text Generation • 18B • Updated • 257 • 1 -
nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16
Image-Text-to-Text • 124B • Updated • 72.2k • 119 -
nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4
Image-Text-to-Text • 66B • Updated • 54.6k • • 266
EXL3 quants of Liquid AI's LFM2.5-24B-A2B, 2.5 to 6.0 bpw, made with exllamav3 1.6.0 for TandemLLM 0.9.0. LFM Open License v1.0.
-
0xBakeer/TandemLLM-LFM2.5-24B-A2B-EXL3-2.5
Text Generation • 4B • Updated • 184 • 1 -
0xBakeer/TandemLLM-LFM2.5-24B-A2B-EXL3-3.0
Text Generation • 5B • Updated • 187 • 1 -
0xBakeer/TandemLLM-LFM2.5-24B-A2B-EXL3-3.5
Text Generation • 6B • Updated • 208 • 1 -
0xBakeer/TandemLLM-LFM2.5-24B-A2B-EXL3-4.0
Text Generation • 6B • Updated • 203 • 1
Weights and drafters of the StairCut paper (doi.org/10.5281/zenodo.23045275). Engine: github.com/0xBakeer/TandemLLM
EXL3 pack of Ornith-1.5-122B at 4.0 bpw, for TandemLLM on DGX Spark. MIT.
EXL3 packs of NVIDIA Nemotron 3.5 Bolt 74B-A9B, 2.0 to 6.0 bpw, for TandemLLM 0.10.0 on DGX Spark. OpenMDW-1.1.
-
0xBakeer/TandemLLM-Nemotron-3.5-Bolt-74B-A9B-EXL3-2.0
Text Generation • 12B • Updated • 286 • 1 -
0xBakeer/TandemLLM-Nemotron-3.5-Bolt-74B-A9B-EXL3-2.5
Text Generation • 14B • Updated • 285 • 1 -
0xBakeer/TandemLLM-Nemotron-3.5-Bolt-74B-A9B-EXL3-3.0
Text Generation • 16B • Updated • 261 • 1 -
0xBakeer/TandemLLM-Nemotron-3.5-Bolt-74B-A9B-EXL3-3.5
Text Generation • 18B • Updated • 273 • 1
Qwen3.8-27B in EXL3 at six sizes, 2.5 to 6.0 bits per weight, and the EXL3 drafters TandemLLM uses for speculative decoding.
-
0xBakeer/TandemLLM-Qwen3.8-27B-EXL3-2.5
Image-Text-to-Text • 6B • Updated • 63 • 1 -
0xBakeer/TandemLLM-Qwen3.8-27B-EXL3-3.0
Image-Text-to-Text • 7B • Updated • 120 • 1 -
0xBakeer/TandemLLM-Qwen3.8-27B-EXL3-3.5
Image-Text-to-Text • 8B • Updated • 50 -
0xBakeer/TandemLLM-Qwen3.8-27B-EXL3-4.0
Image-Text-to-Text • 8B • Updated • 153 • 1
Recursive's Flint for TandemLLM on one DGX Spark: our EXL3 packs at 4.0 (the default) and 6.0 bits a weight, and Recursive's BF16 checkpoint.
EXL3 pack of Ornith-1.5-122B at 4.0 bpw, for TandemLLM on DGX Spark. MIT.
EXL3 packs of NVIDIA Nemotron 3.5 Super VL 120B-A12B at 4.0 and 2.0 bpw, for TandemLLM 0.11.0 on DGX Spark. OpenMDW-1.1.
-
0xBakeer/TandemLLM-Nemotron-3.5-Super-VL-120B-A12B-EXL3-4.0
Text Generation • 32B • Updated • 258 • 1 -
0xBakeer/TandemLLM-Nemotron-3.5-Super-VL-120B-A12B-EXL3-2.0
Text Generation • 18B • Updated • 257 • 1 -
nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16
Image-Text-to-Text • 124B • Updated • 72.2k • 119 -
nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4
Image-Text-to-Text • 66B • Updated • 54.6k • • 266
EXL3 packs of NVIDIA Nemotron 3.5 Bolt 74B-A9B, 2.0 to 6.0 bpw, for TandemLLM 0.10.0 on DGX Spark. OpenMDW-1.1.
-
0xBakeer/TandemLLM-Nemotron-3.5-Bolt-74B-A9B-EXL3-2.0
Text Generation • 12B • Updated • 286 • 1 -
0xBakeer/TandemLLM-Nemotron-3.5-Bolt-74B-A9B-EXL3-2.5
Text Generation • 14B • Updated • 285 • 1 -
0xBakeer/TandemLLM-Nemotron-3.5-Bolt-74B-A9B-EXL3-3.0
Text Generation • 16B • Updated • 261 • 1 -
0xBakeer/TandemLLM-Nemotron-3.5-Bolt-74B-A9B-EXL3-3.5
Text Generation • 18B • Updated • 273 • 1
EXL3 quants of Liquid AI's LFM2.5-24B-A2B, 2.5 to 6.0 bpw, made with exllamav3 1.6.0 for TandemLLM 0.9.0. LFM Open License v1.0.
-
0xBakeer/TandemLLM-LFM2.5-24B-A2B-EXL3-2.5
Text Generation • 4B • Updated • 184 • 1 -
0xBakeer/TandemLLM-LFM2.5-24B-A2B-EXL3-3.0
Text Generation • 5B • Updated • 187 • 1 -
0xBakeer/TandemLLM-LFM2.5-24B-A2B-EXL3-3.5
Text Generation • 6B • Updated • 208 • 1 -
0xBakeer/TandemLLM-LFM2.5-24B-A2B-EXL3-4.0
Text Generation • 6B • Updated • 203 • 1
Qwen3.8-27B in EXL3 at six sizes, 2.5 to 6.0 bits per weight, and the EXL3 drafters TandemLLM uses for speculative decoding.
-
0xBakeer/TandemLLM-Qwen3.8-27B-EXL3-2.5
Image-Text-to-Text • 6B • Updated • 63 • 1 -
0xBakeer/TandemLLM-Qwen3.8-27B-EXL3-3.0
Image-Text-to-Text • 7B • Updated • 120 • 1 -
0xBakeer/TandemLLM-Qwen3.8-27B-EXL3-3.5
Image-Text-to-Text • 8B • Updated • 50 -
0xBakeer/TandemLLM-Qwen3.8-27B-EXL3-4.0
Image-Text-to-Text • 8B • Updated • 153 • 1
Weights and drafters of the StairCut paper (doi.org/10.5281/zenodo.23045275). Engine: github.com/0xBakeer/TandemLLM