Record · corpus.blog/posts/01a0877d-0b2f-70d1-a88c-a0ab1c89c7f1

Benchmarking Qwen 3.6 35B MoE (3B active) on an RTX 3090

Giles' blog · published 24 July 2026

A record is what was published and who pointed at it. The text of the post is not held here: read it at the source, or at its Wayback capture.

Read the post at gilesthomas.com Wayback capture

0blogs citing

First cited: —
Most recent: —
Rank this month: not ranked
External links: 21
Words captured: 9,607

Cited by

No blog we hold has cited this post yet.

Links from this post

Link

Host

https://llama-cpp.com/

llama-cpp.com

https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF

huggingface.co

https://huggingface.co/Qwen/Qwen3.6-35B-A3B

huggingface.co

https://huggingface.co/docs/hub/main/en/gguf

huggingface.co

https://wiki.archlinux.org/title/Llama.cpp

wiki.archlinux.org

https://huggingface.co/Qwen/Qwen3-4B-GGUF

huggingface.co

https://wiki.archlinux.org/title/Talk:Llama.cpp

wiki.archlinux.org

https://mbrenndoerfer.com/writing/repetition-penalties-language-model-generation

mbrenndoerfer.com

https://unsloth.ai/

unsloth.ai

https://www.reddit.com/r/LocalLLaMA/comments/1t49pqu/comment/ok1if2y/

reddit.com

https://x.com/emollick/status/1765136992176644281

x.com

https://www.gutenberg.org/ebooks/1342

gutenberg.org

https://github.com/gpjt/measure-llama-speeds/blob/main/measure_ncmoe_speeds.py

github.com

https://github.com/gpjt/measure-llama-speeds/blob/main/sweep_results/original_vulkan/sweep_results_v1.csv

github.com

https://github.com/ggml-org/llama.cpp

github.com

https://github.com/gpjt/measure-llama-speeds/blob/main/sweep_results/cuda/sweep_results.csv

github.com

https://x.com/emollick

x.com

https://www.oneusefulthing.org/

oneusefulthing.org

https://www.goodreads.com/quotes/400182-have-it-compose-a-poem--a-poem-about-a-haircut

goodreads.com

https://huggingface.co/Qwen/Qwen3.5-397B-A17B

huggingface.co

https://github.com/danveloper/flash-moe

github.com