Qwen3.6-27B-MTP-UD-Q5_K_XL on my 7900XTX goes from 32 t/s to 50-72 t/s depending on the predictability of the task. So, a 1.5x increase on creative tasks up to a 2.2x increase on math.

MTP does not change the quality with the only cost being a few hundred MB extra VRAM usage. You will need to download a gguf model with MTP support to use it.
My parameters:

; Context memory usage  
ctx-size = 65536  
ctk = q8_0  
ctv = q8_0  

; Prompt processing speed  
batch-size = 1024  
ubatch-size = 1024  

; Speculative decoding  
np = 1  
spec-type = draft-mtp  
spec-draft-n-max = 3  

Edit: did some more testing using Unsloth's parameters and with spec-draft-n-max = 6 I can get up to 82 tk/s, a 2.56x increase, on the same math prompt. But this comes at the cost of the creative writing task that now falls below 40 tk/s.
It seems like this should be tweaked depending on the prompt similar to the sampling parameters.

DeepSeek-V4-Pro-Max, the maximum reasoning effort mode of DeepSeek-V4-Pro, significantly advances the knowledge capabilities of open-source models, firmly establishing itself as the best open-source model available today. It achieves top-tier performance in coding benchmarks and significantly bridges the gap with leading closed-source models on reasoning and agentic tasks. Meanwhile, DeepSeek-V4-Flash-Max achieves comparable reasoning performance to the Pro version when given a larger thinking budget, though its smaller parameter scale naturally places it slightly behind on pure knowledge tasks and the most complex agentic workflows.

Both models support one million tokens context length.

Recently made a post about the 35b MOE. Now the dense 27b variant has been released.

The Qwen3.5 models are still the best local models I’ve used, so I’m excited to see how this updated version performs.

Also available on mobile: Checkout link Android & iOS

Long but interesting watch. They talk about his history as a ‘militant vegan’, his current views on effective activism, and if he has any hope left for the future of veganism after he retired.

cross-posted from: https://piefed.zip/c/fosai/p/958141/30b-a3b-glm-4-7-flash-released

Small/fast model with MIT license for local use.

Benchmarks look good for the size. But IMO these smaller models aren’t consistent enough to live up to their promises.

Small/fast model with MIT license for local use.

Benchmarks look good for the size. But IMO these smaller models aren’t consistent enough to live up to their promises.