OpensourceWTF/Beam-1.0-501B-A23B-NVFP4-q8-mlx-stream

Text Generation • 3B • Updated 25 days ago • 444 • 1

Note Beam 1.0 501B-A23B: NVFP4 experts, MLX 8-bit residents, streaming pack for mlx-serve + mlx-stream.

OpensourceWTF/MiMo-V2.6-Flash-MOPD-fp8res-mlx-stream

Text Generation • 6B • Updated Oct 12 • 299

Note MiMo-V2.6-Flash-MOPD, fp8 residents, streaming pack for mlx-serve + mlx-stream.

OpensourceWTF/Muse-Glimmer-30B-MLX-q4

Text Generation • 28B • Updated Aug 10 • 202 • 1

Note Muse-Glimmer-30B text tower, 4-bit affine MLX quantization for MTPLX.

OpensourceWTF/DeepSeek-V4.1-Flash-1017-exl3-2.81bpw

Updated 8 days ago • 487 • 4

Note plain exllamav3 layout (CUDA); the streaming version is in EXL3 streaming packs