impulse

Qwen3 235B A22B

Open weights
API id Qwen3-235B-A22B
Input / 1M
$0.18
at DeepInfra
Output / 1M
$0.54
at DeepInfra
Context window
41K
tokens
Max output
41K
tokens per reply
Hosts
6
6 listings
Released
-
date not published

Pricing

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

TierInput / 1MOutput / 1M
Standard$0.18$0.54

What it costs a month

At DeepInfra list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out$94.50
  • RAG over documents2,000 requests a day, 8K in / 0.6K out$106
  • Coding agent1,000 requests a day, 40K in / 2K out$248
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$562
Use your own numbers

Where to run it

6 hosts serve Qwen3 235B A22B. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
DeepInfraInference
Qwen3-235B-A22B
$0.18 / $0.54
Qwen3-235B-A22B
$0.20 / $0.60
+11% vs headline
Novita AIInference
qwen3-235b-a22b-fp8
$0.20 / $0.80
+30% vs headline
Fireworks AIInference
qwen3-235b-a22b
$0.22 / $0.88
+43% vs headline
GMI CloudInference
Qwen3-235B-A22B-FP8
$0.17 / $1.09
+48% vs headline
HyperbolicInference
Qwen3-235B-A22B
$2.00 / $2.00
+641% 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
Text
Produces
Text
Open weights
License: Apache-2.0
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

No host sells it for less than DeepInfra.

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
    26% less
  • GPT-6 Luna
    OpenAI · on OpenAI
    $0.10 / $0.50
    Has every key capability
    26% less
  • GPT-4o Mini
    OpenAI · on OpenAI
    $0.15 / $0.60
    Has every key capability
    3% less

Open-weight picks

  • Mistral Small 2603
    Mistral AI · on Mistral AI
    $0.15 / $0.60
    Has every key capability
    3% less
  • Qwen3.8 27B
    Alibaba Qwen · on EmpirioLabs AI
    $0.17 / $0.50
    Has every key capability
    6% less
  • Nemotron 3.5 Super VL
    NVIDIA · on Kyma API
    $0.081 / $0.472
    Has every key capability
    34% less
  • Gemma 4 31B IT
    Google · on W&B Inference
    $0.10 / $0.34
    Has every key capability
    41% less

Benchmarks

Scores published on the model card.

  • Aider Polyglot59.6%

Call it

Using DeepInfra's API.

Python
from litellm import completion

response = completion(
    model="deepinfra/Qwen3-235B-A22B",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Qwen3 235B A22B cost?

On DeepInfra, Qwen3 235B A22B costs $0.18 per million input tokens and $0.54 per million output tokens.

Which APIs offer Qwen3 235B A22B?

Qwen3 235B A22B runs on 6 APIs: DeepInfra, Nebius AI Studio, Novita AI, Fireworks AI, GMI Cloud and Hyperbolic.

What is Qwen3 235B A22B's context window?

41K tokens, with up to 41K tokens in one reply.

Is Qwen3 235B A22B being retired?

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

What are cheaper alternatives to Qwen3 235B A22B?

Models that cost less and keep most of what it can do: Claude Haiku 5.5 on Anthropic (26% less), GPT-6 Luna on OpenAI (26% less) and Mistral Small 2603 on Mistral AI (3% less). Savings are on a mix of 3 input tokens to 1 output token.

What is Qwen3 235B A22B's API model id?

Use Qwen3-235B-A22B with DeepInfra. Other hosts use their own ids: qwen3-235b-a22b-fp8 on Novita AI, qwen3-235b-a22b on Fireworks AI, Qwen3-235B-A22B-FP8 on GMI Cloud.

Sources

Raw catalog entry for Qwen3-235B-A22B on DeepInfra
{
  "max_input_tokens": 40960,
  "max_output_tokens": 40960,
  "input_cost_per_token": 1.8e-7,
  "output_cost_per_token": 5.4e-7,
  "litellm_provider": "deepinfra",
  "mode": "chat",
  "supports_tool_choice": true,
  "supports_function_calling": true
}