OpenModelsHQ

Qwen3.6-35B-A3B

Alibaba QwenVisionMoEApache-2.0China Desk

Qwen3.6-35B-A3B is an open-weight vision and language model from Alibaba Qwen with 36B parameters (mixture of experts). At 4-bit it needs about 23.9 GB of memory, so the smallest current device that fits it is the Radeon AI PRO R9700.

Overview

Estimated memory at 4-bit
23.9GB

Includes room for an 8k context and runtime overhead. More context needs more memory.

Total parameters
36B
Mixture of experts, all weights counted
Context window
256K
262,144 tokens
Cheapest API, blended / 1M
$0.0465
₹4.38 · PrimeIntellect · 3:1 input to output
All Alibaba Qwen API prices

Key information

Lab
Alibaba Qwen
Released
15 Apr 2026
Parameters
36B · MoE
Type
Vision and language
Context
262,144 tokens
License
Apache-2.0
HF downloads, 30 days
4,036,202
View all

Will it run on your hardware?

How we estimate
Memory needed vs devices
Qwen3.6-35B-A3B needs about 23.9 GB at 4-bit, 42.1 GB at 8-bit and 79.3 GB at 16-bit. At 4-bit it fits GeForce RTX 5090, DGX Spark, Mac Studio M5 Ultra (256GB). Dashed lines are usable memory on each device; a model fits when its estimate stays within 90% of it. Log scale. 8k context.
Memory needed vs devices: data
Estimate or deviceMemory
Q423.9 GB
Q842.1 GB
BF1679.3 GB
RTX 5060 Ti 16 GB16 GB usable
RTX 5090 32 GB32 GB usable
DGX Spark 128 GB108.8 GB usable
Mac Studio M5 Ultra 256 GB192 GB usable

Single-device memory estimate. Pick a device to update the On your device column.

36 GB usable memory

FormatMemory neededOn your deviceExample hardware
4-bit (Q4)≈ 23.9 GBFits 66%Radeon AI PRO R970032 GB memory
8-bit (Q8)≈ 42.1 GBToo big 117%L40S48 GB memory
16-bit (BF16)≈ 79.3 GBToo big 220%RTX PRO 6000 Blackwell96 GB memory

Fits means the estimate stays within 90% of usable memory. Tight is above 90% and up to 100%. Too big is above 100%. These are memory checks, not speed or compatibility tests.

Mac allowances are 66% of total RAM up to 36 GB, and 75% above. AI boxes allow 85% because their memory is shared with the operating system.

Where to run it

On a Mac

Start with the MLX files, built for Apple Silicon.

lmstudio-community/Qwen3.6-35B-A3B-MLX-4bitCheck your Mac

With GGUF

Choose a file for llama.cpp or a compatible app.

unsloth/Qwen3.6-35B-A3B-GGUFFind compatible hardware

Through an API

PrimeIntellect (fp8) is the cheapest provider we found.

$0.015 / $0.141

Input / output per 1M tokens
₹1.41 / ₹13 at ₹94.29 per $1

Blended: $0.0465 (₹4.38) per 1M. Ranked by a 3:1 mix of input and output tokens.

MLX on Apple SiliconView repository
mlx_vlm.generate --model lmstudio-community/Qwen3.6-35B-A3B-MLX-4bit \
  --prompt "Describe this image." --image image.jpg
GGUF with llama.cppView repository
llama-cli -hf unsloth/Qwen3.6-35B-A3B-GGUF \
  -c 8192 -p "Explain how a transformer works."

Example commands built from the repository ids above. They need a working install of MLX or llama.cpp. We have not run them; check each repository for available files.

Hosted API 10

USD and INR per million tokensAll Alibaba Qwen API prices
Precision
PrimeIntellectCheapest$0.0465₹4.38$0.015₹1.41$0.141₹13262,144FP8
Darkbloom$0.2125₹20$0.05₹4.71$0.70₹66262,144FP4
CoreWeave$0.2975₹28$0.10₹9.43$0.89₹84262,144FP8
AkashML$0.30₹28$0.10₹9.43$0.90₹85262,144FP8
Parasail$0.30₹28$0.10₹9.43$0.90₹85262,144FP8
DeepInfra$0.3125₹29$0.10₹9.43$0.95₹90262,144FP8
DekaLLM$0.325₹31$0.10₹9.43$1.00₹94262,144Not listed
Venice$0.325₹31$0.10₹9.43$1.00₹94256,000FP8
AtlasCloud$0.50₹47$0.25₹24$1.25₹118262,144FP8
SiliconFlow$0.63₹59$0.24₹23$1.80₹170262,144FP8
View on OpenRouterEndpoint sourceSorted by the blended price: 3 parts input to 1 part output. Context limits differ by provider.

INR at ₹94.29 per $1. Prices fetched 14 Nov 2026.

More about this model

Frequently asked questions

How much memory does Qwen3.6-35B-A3B need? At an 8k context, about 23.9 GB at 4-bit, 42.1 GB at 8-bit and 79.3 GB at 16-bit (BF16). A longer context needs more.

Can I run Qwen3.6-35B-A3B on a Mac? The smallest current Mac that fits the 4-bit estimate of 23.9 GB is the Mac mini M5 Pro (48GB). An MLX build is available: lmstudio-community/Qwen3.6-35B-A3B-MLX-4bit.

Does the full context fit? Qwen3.6-35B-A3B supports up to 262,144 tokens of context. Our memory estimates use 8k, so long prompts need more memory than shown.

What does it cost through an API? The cheapest provider we found is PrimeIntellect, at $0.02 per 1M input tokens and $0.14 per 1M output tokens (₹1.41 and ₹13 at 94.29 rupees per dollar). Ranked by a 3:1 mix of input and output tokens.

License in plain English

Apache-2.0. This license permits commercial use, modification and redistribution, with conditions such as keeping the required notices. This summary is not legal advice.

Read Apache-2.0

Read about this license

OMHQ RigBench

Not yet tested. We show no speed or quality score without a measurement.

Revision history and sources