Leaderboard / Alibaba (Qwen)

Qwen3.5 35B-A3B

Name: Qwen3.5 35B-A3B
Price: 0.236 USD
Author: Alibaba (Qwen)

Alibaba (Qwen) · Qwen3.5 · released 2026-02-24 · data checked 2026-11-01

Qwen3.5 35B-A3B is a 35B-parameter Mixture-of-Experts model (3B active per token) from Alibaba (Qwen) (China), released 2026-02-24 under the Apache-2.0 license. It scores 69.9 on the Small LM Index, #9 of 34 medium models. It needs about 22.6 GB of memory at 4-bit with an 8K-token context and costs $0.24 per million tokens through the cheapest API provider.

Small LM Index

69.9

#9 of 34 scored medium models

Likely 68.5–71.3 · from 6 of 6 core benchmarks

Parameters

35B

3B active · MoE

Price per 1M tokens

$0.24

$0.065 in / $0.75 out · blended 3:1

Context window

262K

tokens

Benchmarks

Each bar shows this model's score. The tick marks the best medium model on that benchmark. Sources: 8 reported by Alibaba (Qwen). How we label sources.

MMLU-Pro Lab-reported Knowledge85.3

Class best: Qwen3.6 27B 86.2

GPQA Diamond Lab-reported Knowledge84.2

Class best: Nanbeige4.3 33B 89.5

Humanity's Last Exam Lab-reported Reasoning22.4

Class best: Qwen3.8 27B 30.8

AIME Lab-reported Math · AIME 202691.0

Class best: Ternary Bonsai 2 27B 95.8

LiveCodeBench Lab-reported Coding · LiveCodeBench v674.6

Class best: Qwen3.8 27B 90.3

SWE-bench Verified Lab-reported Coding70.0

Class best: Qwen3.6 27B 77.2

Terminal-Bench Lab-reported Agentic · Terminal-Bench 2.040.5

Class best: Nanbeige4.3 33B 80.9

MMMU-Pro Lab-reported Vision75.1

Class best: Gemma 4 31B 76.9

Specifications

Lab: Alibaba (Qwen) (CN · China)
Architecture: Mixture-of-Experts
Total parameters: 35B
Active parameters: 3B
Size class: Medium
Context window: 262K tokens
Input modalities: text, image, video
Reasoning: Hybrid (thinking on/off)
License: Apache-2.0
Openness: Open weights
Memory to run: ~22.6 GB at 4-bit, 8K context (more sizes)
Hugging Face: 1.7M downloads (30 days) · 1.6K likes
API providers: 7 on OpenRouter
API price: $0.065 in / $0.75 out per 1M (OpenRouter, 2026-11-03)

Weights on Hugging Face ↗Try on OpenRouter ↗

Compare with

Qwen3.5 35B-A3B vs Qwen3.6 35B-A3B (35B, index 72.2)
Qwen3.5 35B-A3B vs MiniCPM-V 4.7 35B-A3B (35.2B, index 54.5)
Qwen3.5 35B-A3B vs Edge0 35B-A3B (preview) (34.7B, index 62.8)
Qwen3.5 35B-A3B vs Seed-OSS 36B Instruct (36.2B, index 60.3)
Qwen3.5 35B-A3B vs Hermes 4.3 36B (36.2B, index 51.4)

Ranked in

Sources

https://www.alibabacloud.com/blog/qwen3-6-35b-a3b-thinking-open-weights-uncompromised-agentic-coding_603043

Something wrong here? Report it.

Run it locally

Uses unsloth/Qwen3.5-35B-A3B-GGUF (Q4_K_M). Swap the tag for another size: BF16, MXFP4, Q3_K_M, Q3_K_S, Q4_K_M, Q4_K_S, Q5_K_M, Q5_K_S….

Ollama

ollama run hf.co/unsloth/Qwen3.5-35B-A3B-GGUF:Q4_K_M

llama.cpp

llama-server -hf unsloth/Qwen3.5-35B-A3B-GGUF:Q4_K_M

Then open http://localhost:8080 for a chat window.

MLX (Mac)

pip install -U mlx-lm
mlx_lm.chat --model mlx-community/Qwen3.5-35B-A3B-OptiQ-4bit

Text only; for images use mlx-vlm.

vLLM (GPU server)

vllm serve Qwen/Qwen3.5-35B-A3B

Full-precision weights, for a server GPU.

Open in LM Studio

Memory needed: weights, context cache and runtime buffers.

Quantization	4K context	32K context	256K context
16-bit F16	71 GB	72 GB	77 GB
8-bit Q8_0	38 GB	39 GB	44 GB
4-bit Q4_K_M	23 GB	23 GB	28 GB

Check it against your GPU, Mac or phone →