tablinum/Moon-Beam-8B-A1B-GGUF-Preview

Text Generation • 8B • Updated about 1 hour ago • 461 • 2

Note 8B-A1B GGUF ladder (Q8_0 to IQ1_M) + MTP draft head. One command: terrabeam run tablinum/Moon-Beam-8B-A1B-GGUF-Preview

tablinum/Moon-Beam-16B-A3B-GGUF

Text Generation • 16B • Updated 1 day ago • 1.3k • 2

Note IQ2 / IQ3 / IQ4 GGUF for llama.cpp - IQ3 recommended

tablinum/Moon-Beam-8B-A1B-MLX-Preview

Text Generation • 8B • Updated about 1 hour ago • 91

Note 8B-A1B MLX 4.6 GB (4-bit experts, 8-bit rest) for Apple Silicon.

tablinum/Moon-Beam-16B-A3B

Text Generation • 16B • Updated 4 days ago • 1.12k

Note BF16 weights (safetensors)

Note The Tablinum-2 engram it is trained to use

tablinum/Moon-Beam-8B-A1B-Preview

Text Generation • 8B • Updated about 1 hour ago • 154 • 1

Note 8B-A1B (Ling-3.0-tiny), BF16 safetensors + MTP layer. Preview: Tablinum-2 library until Tablinum-3.

tablinum/Moon-Beam-8B-A1B-FP8-Preview

Text Generation • 8B • Updated about 1 hour ago • 23

Note 8B-A1B FP8 block-128 for vLLM / SGLang.

tablinum/Moon-Beam-8B-A1B-MLX-8bit-Preview

Text Generation • 8B • Updated about 1 hour ago • 24

Note 8B-A1B MLX 8-bit for Apple Silicon.