LOADING DATASHEET
GROUNDED IN REAL HUMAN CONVERSATIONSYOUR BENCH
Qwen3.5 27B
◆ OPEN
by Alibaba
OPEN WEIGHTSLOCALAPILONG CONTEXTPAIDREASONING
RANKED #192 OF 228 ASSISTANTS · OVERALL #295 OF 6,618 · VIBE SCORE 5.4 · 82 VOICES
The Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference speed and performance. Its overall capabilities are comparable to those of...
HAPPY7.3
9 mentions
SAD7.7
11 mentions
VIBE TRAJECTORY
26 weeks · 134 voices
HOLDING UP
Recent vibe is steady. No clear fade in the crowd.
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 27B once and it's yours. No subscription, no rate limits, works offline.
27B parameters, a Q4 file is about 16.8 GB, full FP16 is about 55 GB
- BARE MINIMUM
Desktop PC with 32 GB DDR5 RAM and RTX 3060 12GB
12 GB VRAM PLUS 32 GB RAM
~$280 USED GPU OR $650Q3_K or Q4_K with partial GPU offloading, ~6 tok/s, 8k context
- SWEET SPOT
Nvidia GeForce RTX 3090 24GB or RTX 4090 24GB
24 GB VRAM
~$700 USED RTX 3090Q4_K_M fully in VRAM, ~35 tok/s, 32k context
- FULL POWER
Dual Nvidia RTX 3090 24GB (48GB) or Mac Studio M2 Ultra (64G
48 GB TO 64 GB UNIFIED MEMORY OR VRAM
~$1400 USED DUAL GPUQ8_0 or full BF16 precision, ~45 tok/s, 128k+ context
LEARN HOW TO SET IT UP
WILL MY BOX RUN QWEN3.5 27B?
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 27B takes, and how good the crowd says it is at that kind of work.
| THE JOB | BARE MINIMUMDesktop PC with 32 GB DDR5 RAM and RTX 3060 12GB | SWEET SPOTNvidia GeForce RTX 3090 24GB or RTX 4090 24GB | FULL POWERDual Nvidia RTX 3090 24GB (48GB) or Mac Studio M2 Ultra (64G |
|---|---|---|---|
| Write an emaila paragraph or two, drafted from a one-line brief | ~83 S★★★★★ | ~14 S★★★★★ | ~11 S★★★★★ |
| Summarize a documenta long report boiled down to the points that matter | ~4.2 MIN★★★★★ | ~43 S★★★★★ | ~33 S★★★★★ |
| Build a websitea small landing page, markup and styles together | ~22 MIN★★★★★ | ~3.8 MIN★★★★★ | ~3 MIN★★★★★ |
| Build a backendan API with routes, storage and tests | ~69 MIN★★★★★ | ~12 MIN★★★★★ | ~9.3 MIN★★★★★ |
| Build a gamea playable browser game in one file | ~42 MIN★★★★★ | ~7.1 MIN★★★★★ | ~5.6 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 27B right now.
66 POSTS · 16 COMMENTS · HACKER NEWS · DEV FORUMS · GITHUB · REDDIT · LEMMY · STACK OVERFLOW · X
PEOPLE LOVE
- No fan mail yet. The crowd is still warming up.
PEOPLE COMPLAIN ABOUT
- Refuses or lectures more than people like (7 mentions)
TOPIC BY TOPIC
- Saying no too muchA SORE SPOT
0 thumbs up · 7 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 · 2026-07-22 · 326 UPVOTESmicrosoft/Fara1.5-27B · Hugging Face
microsoft/Fara1.5-27B · Hugging Face
- HACKER NEWS · 2026-04-20 · 165 UPVOTESWe got 207 tok/s with Qwen3.5-27B on an RTX 3090
We got 207 tok/s with Qwen3.5-27B on an RTX 3090
- REDDIT · 2026-04-07 · 135 UPVOTESI ran Gemma 4 26B vs Qwen 3.5 27B across 18 real local business tests on my RTX 4090. Gemma won 13 to 5.
I ran Gemma 4 26B vs Qwen 3.5 27B across 18 real local business tests on my RTX 4090. Gemma won 13 to 5.
- REDDIT · 2026-07-09 · 96 UPVOTESReasoning-Medical0.1-27B (Qwen3.5-27B medical finetune, claims to surpass MedGemma)
Reasoning-Medical0.1-27B (Qwen3.5-27B medical finetune, claims to surpass MedGemma)
- REDDIT · 2026-04-13 · 92 UPVOTESQwen 3.5 27B/35BA3B Tool Calling Issues: Why It Breaks & How I Fixed It
Qwen 3.5 27B/35BA3B Tool Calling Issues: Why It Breaks & How I Fixed It
- REDDIT · 2026-08-15 · 84 UPVOTESQwen3.8 27B vs Qwen3.6 27B vs Qwen3.5 27B, a slight improvement in oneshotting ability across generations.
Qwen3.8 27B vs Qwen3.6 27B vs Qwen3.5 27B, a slight improvement in oneshotting ability across generations.
- 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 · 2026-07-08 · 76 UPVOTESOpen-source models are closing the coding gap with GPT/Claude/Gemini ~1.5x faster than the frontier is advancing, and on decontaminated…
Open-source models are closing the coding gap with GPT/Claude/Gemini ~1.5x faster than the frontier is advancing, and on decontaminated benchmarks a 27B model already beats Claude Opus 4.8 [live dashboard + analysis]
- 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 · 2026-05-26 · 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 · 2026-03-13 · 41 UPVOTESWhy is Qwen3.5:27b using over 24GB of VRAM?
Why is Qwen3.5:27b using over 24GB of VRAM?