Image-Text-to-Text
Safetensors
GGUF
English
Chinese
qwen3_5
solstice-ai
davidau
davidau-quants
qwen
qwen3.8
qwen3.8-27b
cold-fusion
gain
project-heretic
heretic
uncensored
fable
cot
reasoning
coding
swe-bench
swe-bench-pro
beats-claude-opus-4.6
nvfp4
fp4
blackwell
rtx-5090
vllm
sglang
ultraoptimised
ultraefficient
efficient
arc-challenge
735-arc
882-arc
conversational
8-bit precision
compressed-tensors
fix(mtp): restore complete 15-tensor MTP drafter from DavidAU base (BF16, unquantized)
#1
by SolsticeAI - opened
MTP drafter was 1/15 tensors. Replaces with full MTP module from DavidAU, registers all 15 tensors, adds mtp.* to quantization_config.ignore for raw BF16 loading.
SolsticeAI changed pull request status to merged