Earlier quoted context omitted.
1. Gemma-4 we re-uploaded 4 times - 3 times were 10-20 llama.cpp bug fixes - we had to notify people to upload the correct ones. The 4th is an official Gemma chat template improvement from Google themselves. 2. Qwen3.5 - we shared our 7TB research artifacts showing which layers not to quantize - all provider's quants were under optimized, not broken - ssm_out and ssm_* tensors were the issue - we're now the best in t…
Thanks for all the amazing work Daniel. I remember you guys being late to OH because you were working on weights released the night before - and it's great to see you guys keep up the speed!
Qwen3.6-35B-A3B: Agentic coding power, now open to all
201–210 of 563 posts
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#202Earlier quoted context omitted.
No Bartowski's are more affected - (38% NaN) than ours (22%) - for MiniMax 2.7 see https://www.reddit.com/r/LocalLLaMA/comments/1slk4di/minimax... We already fixed ours. Bart hasn't yet but is still working on it following our findings. blk.61.ffn_down_exps in Q4_K or Q5_K failed - it must be in Q6_K otherwise it overflows. For the others, yes layers in some precision don't work. For eg Qwen3.5 ssm_out must be minimu…
Thanks again, TIL
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#203Earlier quoted context omitted.
It's on the page: Precision Quantization Tag File Size 1-bit UD-IQ1_M 10 GB 2-bit UD-IQ2_XXS 10.8 GB UD-Q2_K_XL 12.3 GB 3-bit UD-IQ3_XXS 13.2 GB UD-Q3_K_XL 16.8 GB 4-bit UD-IQ4_XS 17.7 GB UD-Q4_K_XL 22.4 GB 5-bit UD-Q5_K_XL 26.6 GB 16-bit BF16 69.4 GB
Additional VRAM is needed for context. This model is a MoE model with only 3B active parameters per expert which works well with partial CPU offload. So in practice you can run the -A(N)B models on systems that have a little less VRAM than you need. The more you offload to the CPU the slower it becomes though.
Or is it only layers but that would affect all Experts?
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#204Earlier quoted context omitted.
taps the sign Unified Memory Is A Marketing Gimmeck. Industrial-Scale Inference Servers Do Not Use It.
Industrial Scale Inference is moving towards LPDDR memory (alongside HBM), which is essentially what "Unified Memory" is.
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#205Earlier quoted context omitted.
Obviously going to depend on your definition of "decent". My impression so far is that you will need between 90GB to 100GB of memory to run medium sized (31B dense or ~110B MoE) models with some quantization enabled.
I’m running Gemma4 31B (Q8) on my 2 4090s (48GB) with no problem.
Any tips around your setup running this?
I use lmstudio with default settings and prioritization instead of split.
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#206Earlier quoted context omitted.
Unsloth is great for uploading quants quickly to experiment with, but everyone should know that they almost always revise their quants after testing. If you download the release day quants with a tool that doesn’t automatically check HF for new versions you should check back again in a week to look for updated versions. Some times the launch day quantizations have major problems which leads to early adopters dismissi…
We re-uploaded Gemma4 4 times - 3 times were due to 20 llama.cpp bug fixes, which we helped solve some as well. The 4th is an official Gemma chat template improvement from Google themselves, so these are out of our hands. All providers had to re-fix their uploads, so not just us. For MiniMax 2.7 - there were NaNs, but it wasn't just ours - all quant providers had it - we identified 38% of bartowski's had NaNs. Ours w…
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#207Already quantized/converted into a sane format by Unsloth: https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF
- Why is Qwen's default "quantization" setup "bad" - Who is Unsloth? - Why is his format better? What gains does a better format give? What are the downsides of a bad format? - What is quantization? Granted, I can look up this myself, but I thought I'd ask for the full picture for other readers.
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#208Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#209Earlier quoted context omitted.
We re-uploaded Gemma4 4 times - 3 times were due to 20 llama.cpp bug fixes, which we helped solve some as well. The 4th is an official Gemma chat template improvement from Google themselves, so these are out of our hands. All providers had to re-fix their uploads, so not just us. For MiniMax 2.7 - there were NaNs, but it wasn't just ours - all quant providers had it - we identified 38% of bartowski's had NaNs. Ours w…
Please publish sha256sums of the merged GGUFs in the model descriptions. Otherwise it's hard to tell if the version we have is the latest.
HF also provides SHA256 for eg https://huggingface.co/unsloth/MiniMax-M2.7-GGUF/blob/main/U... is 92986e39a0c0b5f12c2c9b6a811dad59e3317caaf1b7ad5c7f0d7d12abc4a6e8
But agreed it's probs better to place them in a table
Re: Qwen3.6-35B-A3B: Agentic coding power, now open to all
#210Honestly, this is the AI software I actually look forward to seeing. No hype about it being too dangerous to release. No IPO pumping hype. No subscription fees. I am so pumped to try this!