Qwen 3 Coder
Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team.
qwen-3-coderReleased 23 Jul 2025 (Jul 2025)Pricing
List prices at Kyma API, in US dollars per million tokens.
| Tier | Input / 1M | Output / 1M | |
|---|---|---|---|
| Standard | $0.355 | $1.61 | |
| Cached input 62% off input | $0.135 | 62% off input |
What it costs a month
At Kyma API list prices, 30 days.
- Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$208
- RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$176
- Coding agent1,000 requests a day, 40K in / 2K out, cached$338
- Overnight batch20,000 requests a day, 4K in / 0.4K out$1,238
Where to run it
1 host serve Qwen 3 Coder. The comparison is on blended price, 3 parts input to 1 part output.
| Host | In / out per 1M | Model id | |
|---|---|---|---|
Kyma APICloud qwen-3-coder | $0.355 / $1.61 | qwen-3-coder |
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
- Prompt caching
- Open weights
- License: Apache-2.0Runs in about 1152.4 GB of GPU memoryRuntimes: 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 Kyma API.
Smaller Alibaba Qwen models
No cheaper Alibaba Qwen model keeps its key capabilities.
Other frontier labs
- 31% less
- 70% less
- 70% less
- 61% less
Open-weight picks
- 5% less
- 22% less
- 28% less
- 35% less
Model line
Newer and older versions, at each lab's own price.
- Jul 2025Qwen 3 CoderLatest$0.355 / $1.61
- Jul 2025Qwen3 Coder$0.40 / $1.60
Call it
Using Kyma API's API.
from litellm import completion
response = completion(
model="kyma/qwen-3-coder",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Quick answers
How much does Qwen 3 Coder cost?
On Kyma API, Qwen 3 Coder costs $0.355 per million input tokens and $1.61 per million output tokens. Cached input costs $0.135 per million tokens.
Which APIs offer Qwen 3 Coder?
Qwen 3 Coder runs on 1 API: Kyma API.
What is Qwen 3 Coder's context window?
131K tokens, with up to 33K tokens in one reply.
Is Qwen 3 Coder being retired?
No retirement date is scheduled for it on any API we track.
Is there a newer version of Qwen 3 Coder?
No. Qwen 3 Coder is the newest model in its line.
What are cheaper alternatives to Qwen 3 Coder?
Models that cost less and keep most of what it can do: GPT-5.4 Nano on OpenAI (31% less), Claude Haiku 5.5 on Anthropic (70% less) and Inkling Small on DeepInfra (5% less). Savings are on a mix of 3 input tokens to 1 output token.
When was Qwen 3 Coder released?
Alibaba Qwen released it on 23 Jul 2025.
What is Qwen 3 Coder's API model id?
Use qwen-3-coder with Kyma API.
Sources
Raw catalog entry for qwen-3-coder on Kyma API
{
"max_input_tokens": 131072,
"max_output_tokens": 32768,
"input_cost_per_token": 3.546e-7,
"cache_read_input_token_cost": 1.35e-7,
"output_cost_per_token": 0.000001612,
"litellm_provider": "kyma",
"mode": "chat",
"supports_function_calling": true,
"supports_vision": false,
"supports_prompt_caching": true,
"supports_reasoning": false,
"supports_response_schema": true,
"supports_pdf_input": false
}