incoai
/

Qwen3.8-27B-Splash

Text Generation

local-inference

speculative-decoding

4-bit precision

Model card Files Files and versions

Resources

View closed (0)

Can we have a 8bit target ?

#12 opened 1 minute ago by

Request for more smaller ornith1.5-9B splash or any based on Qwen3.5-9B )

#11 opened about 11 hours ago by

DFLASH2 for Qwen3.8 Flash NExt

#10 opened 2 days ago by

the cli help is incomplete

#9 opened 4 days ago by

why not pi?

#8 opened 4 days ago by

More numbers

#7 opened 4 days ago by

Independent M4 Max 64GB benchmark: Splash vs oMLX vs Ollama on Qwen3.8-27B

#6 opened 4 days ago by

Some MCP calls fail

#5 opened 5 days ago by

Why has the context size been reduced to only 128k?

#4 opened 5 days ago by

Splash Monitor - a native macOS GUI for the Splash engine

#3 opened 6 days ago by

Well Done! 👏 - Speed comparison to few other 3.6 27B and 3.8 27B quants

#2 opened 7 days ago by

Testing on the M4 Max 128G. Yes, it’s faster than the alternatives, yet it still isn’t viable for regular work.

#1 opened 7 days ago by