Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

qwen.ai

Post reply on HN
Print view View original

51–60 of 482 posts

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#51

Post by danielhanchen »

benob wrote: post #27
anonzzzies wrote: post #11

I wish that all announcements of models would show what (consumer) hardware you can run this on today, costs and tok/s.

I get ~5 tokens/s on an M4 with 32G of RAM, using: llama-server \ -hf unsloth/Qwen3.6-27B-GGUF:Q4_K_M \ --no-mmproj \ --fit on \ -np 1 \ -c 65536 \ --cache-ram 4096 -ctxcp 2 \ --jinja \ --temp 0.6 \ --top-p 0.95 \ --top-k 20 \ --min-p 0.0 \ --presence-penalty 0.0 \ --repeat-penalty 1.0 \ --reasoning on \ --chat-template-kwargs '{"preserve_thinking": true}' 35B-A3B model is at ~25 t/s. For comparison, on an A100 (~RTX…

We also made some dynamic MLX ones if they help - it might be faster for Macs, but llama-server definitely is improving at a fast pace.

https://huggingface.co/unsloth/Qwen3.6-27B-UD-MLX-4bit

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#52

Post by jauntywundrkind »

anonzzzies wrote: post #11

I wish that all announcements of models would show what (consumer) hardware you can run this on today, costs and tok/s.

I would detest the time/words it takes to hand hold through such a review, of teaching folks the basics about LLM like this.

It's also a section that, with hope, becomes obsolete sometime semi soon-ish.

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#53

Post by danielhanchen »

verdverm wrote: post #21
cassianoleal wrote: post #10

Earlier quoted context omitted.

> when unsloth has quants available https://huggingface.co/unsloth/Qwen3.6-27B-GGUF

That was quick (compared to the 1T Kimi-2.6, not surprising)

Haha :) We had some issues with Kimi-2.6 since it was int4 and we were investigating how to handle it :)

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#56

Post by rubiquity »

amunozo wrote: post #2

A bit skeptical about a 27B model comparable to opus...

You should try it out. I'm incredibly impressed with Qwen 3.5 27B for systems programming work. I use Opus and Sonnet at work and Qwen 3.x at home for fun and barely notice a difference given that systems programming work needs careful guidance for any model currently. I don't try to one shot landing pages or whatever.

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#57

Post by originalvichy »

jjcm wrote: post #37
originalvichy wrote: post #14

Good news! Friendly reminder: wait a couple weeks to judge the ”final” quality of these free models. Many of them suffer from hidden bugs when connected to an inference backend or bad configs that slow them down. The dev community usually takes a week or two to find the most glaring issues. Some of them may require patches to tools like llama.cpp, and some require users to avoid specific default options. Gemma 4 had…

This is probably less likely with this model, as it’s almost certainly a further RL training continuation of 3.5 27b. The bugs with this architecture were worked out when that dropped.

Valuable note!

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#59

Post by genpfault »

spwa4 wrote: post #15

Unsloth quants available: https://unsloth.ai/docs/models/qwen3.6

Getting ~36-33 tok/s (see the "S_TG t/s" column) on a 24GB Radeon RX 7900 XTX using llama.cpp's Vulkan backend:

    $ llama-server --version
    version: 8851 (e365e658f)

    $ llama-batched-bench -hf unsloth/Qwen3.6-27B-GGUF:IQ4_XS -npp 1000,2000,4000,8000,16000,32000 -ntg 128 -npl 1 -c 34000
    |    PP |     TG |    B |   N_KV |   T_PP s | S_PP t/s |   T_TG s | S_TG t/s |      T s |    S t/s |
    |-------|--------|------|--------|----------|----------|----------|----------|----------|----------|
    |  1000 |    128 |    1 |   1128 |    1.529 |   654.11 |    3.470 |    36.89 |    4.999 |   225.67 |
    |  2000 |    128 |    1 |   2128 |    3.064 |   652.75 |    3.498 |    36.59 |    6.562 |   324.30 |
    |  4000 |    128 |    1 |   4128 |    6.180 |   647.29 |    3.535 |    36.21 |    9.715 |   424.92 |
    |  8000 |    128 |    1 |   8128 |   12.477 |   641.16 |    3.582 |    35.73 |   16.059 |   506.12 |
    | 16000 |    128 |    1 |  16128 |   25.849 |   618.98 |    3.667 |    34.91 |   29.516 |   546.42 |
    | 32000 |    128 |    1 |  32128 |   57.201 |   559.43 |    3.825 |    33.47 |   61.026 |   526.47 |

Re: Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

#60

Post by Mr_Eri_Atlov »

Excited to try this, the Qwen 3.6 MoE they just released a week or so back had a noticeable performance bump from 3.5 in a rather short period of time.

For anyone invested in running LLMs at home or on a much more modest budget rig for corporate purposes, Gemma 4 and Qwen 3.6 are some of the most promising models available.

Post reply on HN