LOADING DATASHEET
GROUNDED IN REAL HUMAN CONVERSATIONSYOUR BENCH
Qwen3.5 35B A3B
◆ OPEN
by Alibaba
OPEN WEIGHTSLOCALAPILONG CONTEXTPAIDREASONING
RANKED #97 OF 228 ASSISTANTS · OVERALL #167 OF 6,618 · VIBE SCORE 6.0 · 72 VOICES
The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency. Its overall...
HAPPY9.8
22 mentions
SAD5.2
4 mentions
VIBE TRAJECTORY
26 weeks · 94 voices
AGING WELL
The crowd is warmer now than it was at the start.
HOW TO USE IT
Open weights. You can use a hosted service, or download it and run it yourself, free.
RUN IT ON YOUR OWN MACHINE
Open weights: download Qwen3.5 35B A3B once and it's yours. No subscription, no rate limits, works offline.
35B total parameters with 3B active per token (MoE), a Q4 file is about 20 GB
- BARE MINIMUM
Nvidia RTX 4060 Ti 16GB
16 GB VRAM + 32 GB SYSTEM RAM
~$450Q3_K_M or Q4 with partial RAM offload, ~15 tok/s, 8k context
- SWEET SPOT
Nvidia RTX 3090 24GB
24 GB VRAM
~$750 USEDQ4_K_M fully in VRAM, ~45 tok/s, 32k context
- FULL POWER
Apple Mac Studio M2 Ultra
128 GB UNIFIED MEMORY
~$3500Q8 or BF16, ~65 tok/s, 128k+ context
LEARN HOW TO SET IT UP
WILL MY BOX RUN QWEN3.5 35B A3B?
The three cards above are researched picks. Search the machine you actually have — we only say yes if it should reply at a usable speed, not just load the weights and crawl.
HARDWARE PICKS & PRICES ARE RESEARCHED FROM THE LIVE WEB AND REFRESHED AUTOMATICALLY. TREAT THEM AS BALLPARK, NOT GOSPEL.
WHAT CAN YOU ACTUALLY DO WITH IT?
Everyday jobs on each of those rigs: how long Qwen3.5 35B A3B takes, and how good the crowd says it is at that kind of work.
| THE JOB | BARE MINIMUMNvidia RTX 4060 Ti 16GB | SWEET SPOTNvidia RTX 3090 24GB | FULL POWERApple Mac Studio M2 Ultra |
|---|---|---|---|
| Write an emaila paragraph or two, drafted from a one-line brief | ~33 S | ~11 S | ~8 S |
| Summarize a documenta long report boiled down to the points that matter | ~1.7 MIN★★★★★ | ~33 S★★★★★ | ~23 S★★★★★ |
| Build a websitea small landing page, markup and styles together | ~8.9 MIN★★★★★ | ~3 MIN★★★★★ | ~2.1 MIN★★★★★ |
| Build a backendan API with routes, storage and tests | ~28 MIN★★★★★ | ~9.3 MIN★★★★★ | ~6.4 MIN★★★★★ |
| Build a gamea playable browser game in one file | ~17 MIN★★★★★ | ~5.6 MIN★★★★★ | ~3.8 MIN★★★★★ |
STARS ARE THE COMMUNITY SCORE FOR THAT KIND OF WORK, DOCKED FOR HOW SQUASHED THE WEIGHTS GET AT EACH TIER. TIMES ARE COMPUTED FROM RESEARCHED SPEEDS. BALLPARK, NOT GOSPEL.
WHAT THE INTERNET SAYS
People are mostly frustrated with Qwen3.5 35B A3B right now.
53 POSTS · 19 COMMENTS · DEV FORUMS · X · HACKER NEWS · LEMMY · REDDIT · GITHUB
PEOPLE LOVE
- No fan mail yet. The crowd is still warming up.
PEOPLE COMPLAIN ABOUT
- Refuses or lectures more than people like (4 mentions)
TOPIC BY TOPIC
- Saying no too muchA SORE SPOT
0 thumbs up · 4 thumbs down
LOOKING FOR AN ALTERNATIVE?
Other models the crowd has fully reviewed, starting with multimodal models like this one.
Grok 4.6
Gemini 3.7 Flash> SOURCES (12)
- REDDIT · 2026-07-10 · 161 UPVOTESHas anyone created a "Local LLM Survival Kit"?
Has anyone created a "Local LLM Survival Kit"?
- REDDIT · 2026-02-24 · 80 UPVOTESQwen just released Qwen 3.5 medium model Series: Qwen 3.5 Flash plus 3 more models
Qwen just released Qwen 3.5 medium model Series: Qwen 3.5 Flash plus 3 more models
- REDDIT · 2026-08-04 · 63 UPVOTESDeepseek V4 Flash 2-bit quant is the first model I can run locally that achieves 100% in this SQL benchmark
Deepseek V4 Flash 2-bit quant is the first model I can run locally that achieves 100% in this SQL benchmark
- REDDIT · 2026-03-25 · 58 UPVOTESOpen WebUI “terminal-aware” skills are scary powerful. I made a skill-building workflow that seems to work well for developing them.
Open WebUI “terminal-aware” skills are scary powerful. I made a skill-building workflow that seems to work well for developing them.
- REDDIT · 2026-05-26 · 57 UPVOTESQwen3.5 27B Uncensored Heretic Native MTP Preserved is Out Now With the Full 15 MTPs Preserved and Retained, Available in Safetensors,…
Qwen3.5 27B Uncensored Heretic Native MTP Preserved is Out Now With the Full 15 MTPs Preserved and Retained, Available in Safetensors, GGUFs, NVFP4, NVFP4 GGUFs and GPTQ-Int4 Formats!
- REDDIT · 2026-05-26 · 54 UPVOTESQwen3.5 27B Uncensored Heretic Native MTP Preserved is Out Now With the Full 15 MTPs Preserved and Retained, Available in Safetensors,…
Qwen3.5 27B Uncensored Heretic Native MTP Preserved is Out Now With the Full 15 MTPs Preserved and Retained, Available in Safetensors, GGUFs, NVFP4, NVFP4 GGUFs and GPTQ-Int4 Formats!
- REDDIT · 2026-08-02 · 51 UPVOTESAll Qwen model oneshots: 1109 outputs to look at and compare!
All Qwen model oneshots: 1109 outputs to look at and compare!
- REDDIT · 2026-07-24 · 46 UPVOTESExtened garlic to run Qwen3.5 35B A3B float8 at 55 tok/s on RTX 5060 Ti
Extened garlic to run Qwen3.5 35B A3B float8 at 55 tok/s on RTX 5060 Ti
- REDDIT · 2026-05-13 · 42 UPVOTESBuilt an open-source one-prompt-to-cinematic-reel pipeline on a single GPU — FLUX.2 [klein] for character keyframes, Wan2.2-I2V for…
Built an open-source one-prompt-to-cinematic-reel pipeline on a single GPU — FLUX.2 [klein] for character keyframes, Wan2.2-I2V for animation, vision critic with auto-retry, music + 9-language narration in the same pipeline
- REDDIT · 2026-05-11 · 29 UPVOTESBuilt an open-source one-prompt-to-cinematic-reel pipeline on a single GPU — FLUX.2 [klein] for character keyframes, Wan2.2-I2V for…
Built an open-source one-prompt-to-cinematic-reel pipeline on a single GPU — FLUX.2 [klein] for character keyframes, Wan2.2-I2V for animation, vision critic with auto-retry, music + 9-language narration in the same pipeline
- REDDIT · 2026-03-15 · 26 UPVOTESProfessional-grade local AI on consumer hardware — 80B stable on 44GB mixed VRAM (RTX 5060 Ti ×2 + RTX 3060) for under €800 total. Full…
Professional-grade local AI on consumer hardware — 80B stable on 44GB mixed VRAM (RTX 5060 Ti ×2 + RTX 3060) for under €800 total. Full compatibility matrix included.
- REDDIT · 2026-02-28 · 18 UPVOTESHas anyone got qwen3.5 to work with ollama?
Has anyone got qwen3.5 to work with ollama?