impulse

Qwen3 Coder 480B A35B Instruct

Open weights
API id qwen3-coder-480b-a35b-instruct-maas
Amazon Bedrock retires it on 30 Mar 2027 (in 132 days). 7 hosts keep serving it.
Input / 1M
$0.22
at Vertex AI
Output / 1M
$1.80
at Vertex AI
Context window
262K
tokens
Max output
33K
tokens per reply
Hosts
8
8 listings
Released
-
date not published

Pricing

List prices at Vertex AI, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.22$1.80
Cached input
90% off input
$0.022
Batch
50% off
$0.11$0.90

What it costs a month

At Vertex AI list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$183
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$123
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$206
  • Overnight batch20,000 requests a day, 4K in / 0.4K out, batch$480
Use your own numbers

Where to run it

8 hosts serve Qwen3 Coder 480B A35B Instruct. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
qwen3-coder-480b-a35b-instruct-maas
$0.22 / $1.80
Novita AIInference
qwen3-coder-480b-a35b-instruct
$0.38 / $1.55
+9% vs headline
DeepInfraInference
Qwen3-Coder-480B-A35B-Instruct
$0.40 / $1.60
+14% vs headline
Qwen3-Coder-480B-A35B-Instruct-FP8
$0.45 / $1.80
+28% vs headline
Mantle
qwen.qwen3-coder-480b-a35b-instruct
$0.45 / $1.80
+28% vs headline
Retires 30 Mar 2027
Fireworks AIInference
qwen3-coder-480b-a35b-instruct
$0.45 / $1.80
+28% vs headline
GMI CloudInference
Qwen3-Coder-480B-A35B-Instruct-FP8
$0.90 / $4.50
+193% vs headline
Together AIInference
Qwen3-Coder-480B-A35B-Instruct-FP8
$2.00 / $2.00
+225% 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 Vertex AI.

Smaller Alibaba Qwen models

No cheaper Alibaba Qwen model keeps its key capabilities.

Other frontier labs

  • GPT-5.4 Nano
    OpenAI · on OpenAI
    $0.20 / $1.25
    Has every key capability
    25% less
  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    67% less
  • GPT-6 Luna
    OpenAI · on OpenAI
    $0.10 / $0.50
    Has every key capability
    67% less
  • GPT-4o Mini
    OpenAI · on OpenAI
    $0.15 / $0.60
    Has every key capability
    57% less

Open-weight picks

  • DeepSeek Flash
    DeepSeek · on DeepSeek
    $0.30 / $1.20
    Has every key capability
    15% less
  • Longcat 2.0
    Independent · on OpenCode Go
    $0.30 / $1.20
    Has every key capability
    15% less
  • Qwen3.7 Plus
    Alibaba Qwen · on Replicate
    $0.276 / $1.10
    Has every key capability
    22% less
  • GLM 4 6v
    Zhipu AI · on Novita AI
    $0.30 / $0.90
    Has every key capability
    27% less

Call it

Using Vertex AI's API.

Python
from litellm import completion

response = completion(
    model="vertex_ai/qwen3-coder-480b-a35b-instruct-maas",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Qwen3 Coder 480B A35B Instruct cost?

On Vertex AI, Qwen3 Coder 480B A35B Instruct costs $0.22 per million input tokens and $1.80 per million output tokens. Cached input costs $0.022 per million tokens.

Which APIs offer Qwen3 Coder 480B A35B Instruct?

Qwen3 Coder 480B A35B Instruct runs on 8 APIs: Vertex AI, Novita AI, DeepInfra, Tensormesh, Amazon Bedrock, Fireworks AI, GMI Cloud and Together AI.

What is Qwen3 Coder 480B A35B Instruct's context window?

262K tokens, with up to 33K tokens in one reply.

Is Qwen3 Coder 480B A35B Instruct being retired?

Amazon Bedrock retires it on 30 Mar 2027. 7 hosts keep serving it.

What are cheaper alternatives to Qwen3 Coder 480B A35B Instruct?

Models that cost less and keep most of what it can do: GPT-5.4 Nano on OpenAI (25% less), Claude Haiku 5.5 on Anthropic (67% less) and DeepSeek Flash on DeepSeek (15% less). Savings are on a mix of 3 input tokens to 1 output token.

What is Qwen3 Coder 480B A35B Instruct's API model id?

Use qwen3-coder-480b-a35b-instruct-maas with Vertex AI. Other hosts use their own ids: qwen3-coder-480b-a35b-instruct on Novita AI, Qwen3-Coder-480B-A35B-Instruct on DeepInfra, Qwen3-Coder-480B-A35B-Instruct-FP8 on Tensormesh, qwen.qwen3-coder-480b-a35b-instruct on Amazon Bedrock.

Sources

Raw catalog entry for qwen3-coder-480b-a35b-instruct-maas on Vertex AI
{
  "cache_read_input_token_cost": 2.2e-8,
  "deprecation_date": "2026-10-21",
  "input_cost_per_token": 2.2e-7,
  "input_cost_per_token_batches": 1.1e-7,
  "litellm_provider": "vertex_ai-qwen_models",
  "max_input_tokens": 262144,
  "max_output_tokens": 32768,
  "mode": "chat",
  "output_cost_per_token": 0.0000018,
  "output_cost_per_token_batches": 9e-7,
  "source": "https://cloud.google.com/gemini-enterprise-agent-platform/generative-ai/pricing",
  "supported_regions": [
    "global"
  ],
  "supports_function_calling": true,
  "supports_pdf_input": false,
  "supports_tool_choice": true
}