What is Qwen3.5-35B-A3B good for?

Text, image and video tasks on a sparse model with 3B active of 35B parameters, the open backbone behind Qwen3.5-Flash. Apache 2.0 weights, on a single consumer GPU.

When is Qwen3.5-35B-A3B not a good fit?

Audio input, image output or built-in web search, which the hosted Qwen3.5-Flash adds. Alibaba's page notes the full context can run out of memory and recommends a smaller window if it does.

What is the cheapest way to run Qwen3.5-35B-A3B?

Hosted, unless you push serious volume. Alibaba Cloud charges $0.25 in / $2.00 out per 1M tokens. The cheapest rental that fits is 2x RTX 4070 at $144 a month, which costs the same as about 266M tokens a month on that API.

How does Qwen3.5-35B-A3B compare with Qwen3.6-35B-A3B?

Alibaba says Qwen3.6-35B-A3B keeps the Qwen3.5-35B-A3B layout and puts its gains into agentic coding, handling frontend work and repository-level reasoning more fluently, and adds an option to carry reasoning context across turns. Same 262K-token context and text, image and video input. Qwen3.5-35B-A3B lists at 34% less per 1M input tokens and 11% less per 1M output tokens than Qwen3.6-35B-A3B. It came out 2 months earlier.

Can I self-host Qwen3.5-35B-A3B?

Yes. The weights are Apache 2.0 licensed. At 4-bit it needs about 22 GB of GPU memory, which starts at roughly $144 a month on the cheapest rental that fits.

Open weights · Apache 2.0

Qwen3.5-35B-A3B

A 35B MoE with 3B active across 256 experts, the open-weight backbone behind Qwen3.5-Flash with native multimodal capability.

Jump to: Specs Pricing Self-hosting Similar models FAQ More from Alibaba Cloud

Visit website Compare vs other LLMs →

List price via Alibaba Cloud

Input: $0.25 /1M tokens
Output: $2.00 /1M tokens

Cheapest: $0.06 / $0.46 per 1M tokens via EmpirioLabs AI

4 providers

Key Specifications

Context window: 262K tokens
Max output: 66K tokens
Released: Feb 2026
Parameters: 35B, 3B active
Inputs: Text, image, video

Capabilities Show details: Reasoning Thinks before it answers, always on or as a switchable mode. Function calling Connect to external tools, APIs, and systems. Structured output Return responses in structured formats like JSON.

Hosted API pricing

Input tokens M Output tokens M

Provider	Input /1M tokens	Output /1M tokens	Cached input /1M	Cost at 10M in + 2M out
EmpirioLabs AI Cheapest	$0.06	$0.46		$1.52	View
Alibaba Cloud Creator	$0.25	$2.00		$6.50	View
Novita	$0.25	$2.00		$6.50	View
EcoHash	$0.40	$0.40	$0.20	$4.80	View

Heads up: Base-tier, on-demand rates per 1M tokens; cached, batch and long-context tiers excluded. A provider may serve a shorter context or a quantized build than the creator's release. Verify before provisioning. More on how we price.

Estimated cost to self-host

Qwen3.5-35B-A3B needs about 22 GB of GPU memory at 4-bit with 32K context. The cheapest rental that fits is 2× RTX 4070 at about $144 a month.

That costs the same as roughly 266M tokens a month on Alibaba Cloud's API. Self-hosting is more expensive below that volume.

Single node ≤ 8

Precision	Memory	Cheapest fit (32K context)(Full 262K context)	Cost /mo	Break-even vs API
4-bitINT4 / FP4	22 GB24 GB	2× RTX 4070	$144$158	266M tokens /mo292M tokens /mo
8-bitFP8 / INT8	38 GB42 GB	4× RTX 3060	$173$173	319M tokens /mo319M tokens /mo
16-bitFP16 / BF16	73 GB77 GB	8× RTX 3060	$346$346	638M tokens /mo638M tokens /mo

Estimates based on median on-demand rates for Nvidia GPUs. Memory is weights plus KV cache for one request, using FP8 KV cache where supported and FP16/BF16 otherwise. Break-even assumes a 5:1 input-to-output ratio. No guarantee of runtime support, usable performance, or that a matching quantized build exists. How we estimate costs.

Similarly priced models

The models nearest Qwen3.5-35B-A3B by blended rate, each at its own cheapest provider.

Model	Blended /1M	Input /1M	Output /1M	Context	Cutoff	vs Qwen3.5-35B-A3B
Llama 3.2 3B Meta	$0.0983	$0.05	$0.34	128K	Dec 2023	−22%
Qwen3-30B-A3B FP8 Alibaba Cloud	$0.0983	$0.05	$0.34	33K		−22%
Ministral 3 3B Mistral	$0.10	$0.10	$0.10	256K		−21%
GLM-5.3-Flash Z.AI	$0.1083	$0.08	$0.25	1M		−14%
GPT-5 Nano OpenAI	$0.1083	$0.05	$0.40	400K	May 2024	−14%
Qwen3.5-35B-A3B This model Alibaba Cloud	$0.1267	$0.06	$0.46	262K
Qwen3.6-35B-A3B Alibaba Cloud	$0.1283	$0.07	$0.42	262K		+1%
Gemma 3 27B Instruct Google Cloud	$0.1333	$0.10	$0.30	128K	Aug 2024	+5%
Gemma 4 26B A4B Google Cloud	$0.1333	$0.10	$0.30	256K	Jan 2025	+5%
Qwen3-32B FP8 Alibaba Cloud	$0.1333	$0.10	$0.30	33K		+5%
Gemini 2.5 Flash-Lite Google Cloud	$0.15	$0.10	$0.40	1M	Jan 2025	+18%

Prices are USD per 1M tokens at each model's cheapest listed provider. Blended is the cost of 10M input plus 2M output tokens, spread over the 12M.

Runpod

Spin Up a GPU in Seconds - No Setup Required

Deploy AI Models Globally in 30+ Regions

Scale from 0 to 1000 GPUs on Demand

PyTorch, Tensorflow & Docker Ready

Learn more

Frequently Asked Questions

What is Qwen3.5-35B-A3B good for?: Text, image and video tasks on a sparse model with 3B active of 35B parameters, the open backbone behind Qwen3.5-Flash. Apache 2.0 weights, on a single consumer GPU.
When is Qwen3.5-35B-A3B not a good fit?: Audio input, image output or built-in web search, which the hosted Qwen3.5-Flash adds. Alibaba's page notes the full context can run out of memory and recommends a smaller window if it does.
What is the cheapest way to run Qwen3.5-35B-A3B?: Hosted, unless you push serious volume. Alibaba Cloud charges $0.25 in / $2.00 out per 1M tokens. The cheapest rental that fits is 2x RTX 4070 at $144 a month, which costs the same as about 266M tokens a month on that API.
How does Qwen3.5-35B-A3B compare with Qwen3.6-35B-A3B?: Alibaba says Qwen3.6-35B-A3B keeps the Qwen3.5-35B-A3B layout and puts its gains into agentic coding, handling frontend work and repository-level reasoning more fluently, and adds an option to carry reasoning context across turns. Same 262K-token context and text, image and video input. Qwen3.5-35B-A3B lists at 34% less per 1M input tokens and 11% less per 1M output tokens than Qwen3.6-35B-A3B. It came out 2 months earlier.
Can I self-host Qwen3.5-35B-A3B?: Yes. The weights are Apache 2.0 licensed. At 4-bit it needs about 22 GB of GPU memory, which starts at roughly $144 a month on the cheapest rental that fits.

More from Alibaba Cloud

Model	Context	Input /1M	Output /1M
Qwen3.6-35B-A3B Open weights	262K	$0.07	$0.42
Qwen3-32B FP8 Open weights	33K	$0.10	$0.30
Qwen3.5-Flash	1M	$0.10	$0.40
Qwen3-30B-A3B FP8 Open weights	33K	$0.05	$0.34
Qwen3-235B-A22B Instruct (2507) Open weights	262K	$0.09	$0.58
Qwen3-Coder-30B-A3B Open weights	262K	$0.06	$0.25
Qwen3.5-27B Open weights	262K	$0.09	$0.69
Qwen3.8-Flash	1M	$0.15	$0.47