impulse

Qwen3 Coder 30B A3B Instruct

Open weights

Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use.

API id qwen3-coder-30b-a3b-instructReleased 31 Jul 2025 (Jul 2025)
Amazon Bedrock retires it on 30 Mar 2027 (in 132 days). 3 hosts keep serving it.
Input / 1M
$0.07
at Novita AI
Output / 1M
$0.27
at Novita AI
Context window
160K
tokens
Max output
33K
tokens per reply
Hosts
4
14 listings
Released
Jul 2025
31 Jul 2025

Pricing

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

TierInput / 1MOutput / 1M
Standard$0.07$0.27

What it costs a month

At Novita AI list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out$41.25
  • RAG over documents2,000 requests a day, 8K in / 0.6K out$43.32
  • Coding agent1,000 requests a day, 40K in / 2K out$100
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$233
Use your own numbers

Where to run it

4 hosts serve Qwen3 Coder 30B A3B Instruct. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
Novita AIInference
qwen3-coder-30b-a3b-instruct
$0.07 / $0.27
Mantle
qwen.qwen3-coder-30b-a3b-instruct
$0.15 / $0.60
+119% vs headline
Retires 30 Mar 2027
Fireworks AIInference
qwen3-coder-30b-a3b-instruct
$0.15 / $0.60
+119% vs headline
qwen3-coder-30b-a3b-instruct
$0.20 / $0.80
+192% vs headline
Qwen3-Coder-30B-A3B-Instruct-GGUF
-
Show 9 more listings: regions, snapshots and variants
Amazon Bedrock
ap-southeast-2 · Mantle
qwen.qwen3-coder-30b-a3b-instruct
$0.154 / $0.618
+125% vs headline
Retires 30 Mar 2027
ap-southeast-3 · Mantle
qwen.qwen3-coder-30b-a3b-instruct
$0.16 / $0.62
+129% vs headline
Retires 30 Mar 2027
eu-west-1 · Mantle
qwen.qwen3-coder-30b-a3b-instruct
$0.18 / $0.70
+158% vs headline
Retires 30 Mar 2027
ap-south-1 · Mantle
qwen.qwen3-coder-30b-a3b-instruct
$0.18 / $0.71
+160% vs headline
Retires 30 Mar 2027
ap-northeast-1 · Mantle
qwen.qwen3-coder-30b-a3b-instruct
$0.18 / $0.73
+165% vs headline
Retires 30 Mar 2027
sa-east-1 · Mantle
qwen.qwen3-coder-30b-a3b-instruct
$0.18 / $0.73
+165% vs headline
Retires 30 Mar 2027
eu-central-1 · Mantle
qwen.qwen3-coder-30b-a3b-instruct
$0.20 / $0.79
+190% vs headline
Retires 30 Mar 2027
eu-south-1 · Mantle
qwen.qwen3-coder-30b-a3b-instruct
$0.20 / $0.79
+190% vs headline
Retires 30 Mar 2027
eu-west-2 · Mantle
qwen.qwen3-coder-30b-a3b-instruct
$0.23 / $0.93
+238% vs headline
Retires 30 Mar 2027

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
Features
Parallel tool calls
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 Novita AI.

Smaller Alibaba Qwen models

No cheaper Alibaba Qwen model keeps its key capabilities.

Other frontier labs

No cheaper match from another frontier lab.

Open-weight picks

  • MiniMax M3.1 Flash
    MiniMax · on Refuseless
    $0.05 / $0.25
    Has every key capability
    17% less
  • Qwen3.5 9B
    Alibaba Qwen · on EmpirioLabs AI
    $0.09 / $0.13
    Has every key capability
    17% less
  • Ministral 3B 2512
    Mistral AI · on Mistral AI
    $0.10 / $0.10
    Has every key capability
    17% less
  • GLM 5.3 Flash
    Zhipu AI · on CLF AI Gateway
    $0.06 / $0.20
    Has every key capability
    21% less

Call it

Using Novita AI's API.

Python
from litellm import completion

response = completion(
    model="novita/qwen3-coder-30b-a3b-instruct",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Qwen3 Coder 30B A3B Instruct cost?

On Novita AI, Qwen3 Coder 30B A3B Instruct costs $0.07 per million input tokens and $0.27 per million output tokens.

Which APIs offer Qwen3 Coder 30B A3B Instruct?

Qwen3 Coder 30B A3B Instruct runs on 4 APIs: Novita AI, Amazon Bedrock, Fireworks AI and Scaleway.

What is Qwen3 Coder 30B A3B Instruct's context window?

160K tokens, with up to 33K tokens in one reply.

Is Qwen3 Coder 30B A3B Instruct being retired?

Amazon Bedrock retires it on 30 Mar 2027. 3 hosts keep serving it.

What are cheaper alternatives to Qwen3 Coder 30B A3B Instruct?

Models that cost less and keep most of what it can do: MiniMax M3.1 Flash on Refuseless (17% less). Savings are on a mix of 3 input tokens to 1 output token.

When was Qwen3 Coder 30B A3B Instruct released?

Alibaba Qwen released it on 31 Jul 2025.

What is Qwen3 Coder 30B A3B Instruct's API model id?

Use qwen3-coder-30b-a3b-instruct with Novita AI. Other hosts use their own ids: qwen.qwen3-coder-30b-a3b-instruct on Amazon Bedrock.

Sources

Raw catalog entry for qwen3-coder-30b-a3b-instruct on Novita AI
{
  "litellm_provider": "novita",
  "mode": "chat",
  "input_cost_per_token": 7e-8,
  "output_cost_per_token": 2.7e-7,
  "max_input_tokens": 160000,
  "max_output_tokens": 32768,
  "supports_function_calling": true,
  "supports_parallel_function_calling": true,
  "supports_tool_choice": true,
  "supports_system_messages": true,
  "supports_response_schema": true
}