impulse

Qwen3.5 35B A3B

Open weights

The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency.

API id Qwen3.5-35B-A3BReleased 25 Feb 2026 (Feb 2026)
Input / 1M
$0.14
at DeepInfra
Output / 1M
$1.00
at DeepInfra
Context window
262K
tokens
Max output
262K
tokens per reply
Hosts
5
5 listings
Released
Feb 2026
25 Feb 2026

Pricing

List prices at DeepInfra, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.14$1.00
Cached input
64% off input
$0.05

What it costs a month

At DeepInfra list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$109
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$81.60
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$152
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$576
Use your own numbers

Where to run it

5 hosts serve Qwen3.5 35B A3B. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
qwen3-5-35b-a3b
$0.057 / $0.459
−56% vs headline
DeepInfraInference
Qwen3.5-35B-A3B
$0.14 / $1.00
W&B InferenceInference
Qwen3.5-35B-A3B
$0.25 / $1.25
+41% vs headline
GMI CloudInference
Qwen3.5-35B-A3B
$0.25 / $2.00
+94% vs headline
Novita AIInference
qwen3.5-35b-a3b
$0.25 / $2.00
+94% vs headline

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
TextImages
Produces
Text
Features
Prompt caching
Open weights
License: Apache-2.0
Runs in about 86.3 GB of GPU memory
Runtimes: Ollama, SGLang, llama.cpp, vLLM

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

  • EmpirioLabs AI
    qwen3-5-35b-a3b
    $0.057 / $0.459
    56% less

Smaller Alibaba Qwen models

No cheaper Alibaba Qwen model keeps its key capabilities.

Other frontier labs

  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    44% less
  • GPT-6 Luna
    OpenAI · on OpenAI
    $0.10 / $0.50
    Has every key capability
    44% less
  • GPT-4o Mini
    OpenAI · on OpenAI
    $0.15 / $0.60
    Lacks reasoning
    26% less

Open-weight picks

  • Muse Glimmer 30B
    Meta · on EmpirioLabs AI
    $0.20 / $0.80
    Has every key capability
    1% less
  • Deepseekv4.1 Flash
    DeepSeek · on Sarvam
    $0.149 / $0.585
    Has every key capability
    27% less
  • Qwen3.8 27B
    Alibaba Qwen · on EmpirioLabs AI
    $0.17 / $0.50
    Has every key capability
    29% less
  • Nemotron 3.5 Super VL
    NVIDIA · on Kyma API
    $0.081 / $0.472
    Has every key capability
    50% less

Model line

Newer and older versions, at each lab's own price.

  1. Apr 2026Qwen3.6 35B A3BLatest$0.10 / $0.89
  2. Feb 2026Qwen3.5 35B A3BThis page$0.14 / $1.00

Call it

Using DeepInfra's API.

Python
from litellm import completion

response = completion(
    model="deepinfra/Qwen3.5-35B-A3B",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Qwen3.5 35B A3B cost?

On DeepInfra, Qwen3.5 35B A3B costs $0.14 per million input tokens and $1.00 per million output tokens. Cached input costs $0.05 per million tokens. The cheapest host is EmpirioLabs AI, at $0.057 per million input tokens and $0.459 per million output tokens.

Which APIs offer Qwen3.5 35B A3B?

Qwen3.5 35B A3B runs on 5 APIs: EmpirioLabs AI, DeepInfra, W&B Inference, GMI Cloud and Novita AI.

What is Qwen3.5 35B A3B's context window?

262K tokens, with up to 262K tokens in one reply.

Is Qwen3.5 35B A3B being retired?

No retirement date is scheduled for it on any API we track.

Is there a newer version of Qwen3.5 35B A3B?

Yes. Qwen3.6 35B A3B is the newest model in this line, released on 27 Apr 2026.

What are cheaper alternatives to Qwen3.5 35B A3B?

Models that cost less and keep most of what it can do: Claude Haiku 5.5 on Anthropic (44% less), GPT-6 Luna on OpenAI (44% less) and Muse Glimmer 30B on EmpirioLabs AI (1% less). Qwen3.5 35B A3B itself costs 56% less on EmpirioLabs AI. Savings are on a mix of 3 input tokens to 1 output token.

When was Qwen3.5 35B A3B released?

Alibaba Qwen released it on 25 Feb 2026.

What is Qwen3.5 35B A3B's API model id?

Use Qwen3.5-35B-A3B with DeepInfra. Other hosts use their own ids: qwen3-5-35b-a3b on EmpirioLabs AI, qwen3.5-35b-a3b on Novita AI.

Sources

Raw catalog entry for Qwen3.5-35B-A3B on DeepInfra
{
  "max_output_tokens": 262144,
  "max_input_tokens": 262144,
  "input_cost_per_token": 1.4e-7,
  "output_cost_per_token": 0.000001,
  "cache_read_input_token_cost": 5e-8,
  "supports_prompt_caching": true,
  "litellm_provider": "deepinfra",
  "mode": "chat",
  "supports_function_calling": true,
  "supports_tool_choice": true,
  "supports_response_schema": true,
  "supports_reasoning": true,
  "supports_vision": true,
  "source": "https://deepinfra.com/pricing"
}