the AI bench
VERIFIED JULY 2026
All models

MODEL · ALIBABA · 35B TOTAL / 3B ACTIVE

Qwen 3.6-35B-A3B

Alibaba's post-3.5 refresh specifically targeting agentic coding — claims to beat dense Qwen 3.5 27B and Gemma 4 31B on coding + reasoning at the same active-param budget. Fully multimodal.

License: Apache 2.0 · Context: 262K native · Released: April 16, 2026

The decision in five lines

The call
Buy — for chat
Best for
chat · agents
Runs on
16 hardware picks fit (cheapest: Minisforum UM890 Pro · $463)
Watch out
When the proven-longest-track-record MoE matters more than the latest — Qwen 3.5 35B-A3B is the equally-Apache, equally-24GB equivalent with more community mileage.
Evidence
Estimated · last verified July 2026

35B total
PARAMETERS
MOE
TYPE
262K
CONTEXT
~17 GB
VRAM AT Q4

Where we recommend this

Every tier slot in the planner where this model is a top or alternate pick. Pulled live from planner.js — when the planner refreshes, this table stays current.

CHAT ·
Qwen 3.6-35B-A3B (the fast daily driver)3B active per token keeps it snappy on unified memory — the reason to run it over 70B dense here is speed, not capability. Step up to 70B/122B for hard reasoning.
CHAT · TOP
Qwen 3.6-35B-A3B3B-active MoE; strong function calling; 262K native context extensible to 1M.
AGENTS · TOP
Qwen 3.6-35B-A3BLatest Qwen MoE; strong function calling; realistic on 24GB+ VRAM or Mac 48GB+ — the local agentic top pick.

The call

Alibaba's post-3.5 refresh specifically targeting agentic coding — claims to beat dense Qwen 3.5 27B and Gemma 4 31B on coding + reasoning at the same active-param budget. Fully multimodal.

When not to use: When the proven-longest-track-record MoE matters more than the latest — Qwen 3.5 35B-A3B is the equally-Apache, equally-24GB equivalent with more community mileage. Otherwise 3.6-35B-A3B is the newer default.

Runner notes

GGUF (`unsloth/Qwen3.6-35B-A3B-GGUF`), MLX, and Ollama tags are all available as of mid-2026; `qwen3.6:35b` is the A3B MoE tag. Q4 fits a single 24 GB rig.

License
Apache 2.0
Released
April 16, 2026
Maker
Alibaba

Hardware that fits

Every hardware pick whose memory fits this model at the quant we recommend. Sorted cheapest-first — the top row is your best-value fit. Click through for the full buyer’s guide.

Next step

Find-by-model — see what hardware runs this