Qwen3-Coder 30B-A3B Instruct

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. Built on the Qwen3 architecture, it supports a native context length of 256K tokens (extendable to 1M with Yarn) and performs strongly in tasks involving function calls, browser use, and structured code completion.

This model is optimized for instruction-following without “thinking mode”, and integrates well with OpenAI-compatible tool-use formats.

alibaba/qwen3-coder-30b-a3b-instruct

Organization: Alibaba
Family: qwen
Providers: 12
Context: 262,144
Output limit: 65,536
Knowledge: 2025-04
Release: 2025-04
Updated: 2025-04
Weights: Open
Input: text
Output: text
Capabilities: Tools, Temperature

Providers

Compare every cataloged offer for this model, including its provider, route, token limits, price, and capabilities.

15 offers · USD per million tokens

Provider	Model ID	Context	Output	Input / output · 1M	Reasoning	Tools	Structured	Details
Amazon Bedrock	qwen.qwen3-coder-30b-a3b-v1:0	262,144	131,072	$0.15 / $0.60	No	Yes	Yes	View
Cortecs	qwen3-coder-30b-a3b-instruct	262,144	262,144	$0.067 / $0.245	No	Yes	Yes	View
DevPass (LLM Gateway)	qwen3-coder-30b-a3b-instruct	262,000	65,536	$0.07 / $0.27	No	Yes	—	View
Eden AI	qwen/qwen3-coder-30b-a3b-instruct	262,144	65,536	$0.45 / $2.25	No	Yes	Yes	View
GreenPT	qwen3-coder-30b-a3b-instruct	128,000	32,768	$0.285 / $1.083	No	Yes	—	View
Hugging Face	Qwen/Qwen3-Coder-30B-A3B-Instruct	262,144	65,536	$0.07 / $0.26	No	Yes	Yes	View
Kilo Gateway	qwen/qwen3-coder-30b-a3b-instruct	262,144	235,929	$0.2925 / $1.4625	No	Yes	Yes	View
LLM Gateway	novita/qwen3-coder-30b-a3b-instruct	160,000	32,768	$0.07 / $0.27	No	Yes	No	View
NanoGPT	qwen/qwen3-coder-30b-a3b-instruct	128,000	65,536	$0.10 / $0.40	No	Yes	Yes	View
Alibabavia OpenRouter	qwen/qwen3-coder-30b-a3b-instruct unknown	262,144	65,536	$0.2925 / $1.4625	No	Yes	No	View
Amazon Bedrockvia OpenRouter	qwen/qwen3-coder-30b-a3b-instruct unknown	0	0	$0.15 / $0.60	No	Yes	No	View
Novitavia OpenRouter	qwen/qwen3-coder-30b-a3b-instruct fp8	160,000	32,768	$0.07 / $0.27	No	Yes	No	View
SiliconFlowvia OpenRouter	qwen/qwen3-coder-30b-a3b-instruct fp8	262,144	235,929	$0.07 / $0.28	No	Yes	Yes	View
Pendra	qwen3-coder:30b	262,144	65,536	$0.00 / $0.00	No	Yes	—	View
Scaleway	qwen3-coder-30b-a3b-instruct	128,000	32,768	$0.20 / $0.80	No	Yes	—	View

Prices and limits apply to each offer. A dash means the value is not provided. OpenRouter offers show the hosting provider and routing channel separately.

Pricing

Compare token rates and cache charges across available routes and hosting configurations.

USD per million tokens

Input from$0.00Per 1M input tokens

Output from$0.00Per 1M output tokens

Available offers15Across 12 providers

Lowest listed input and output prices across the offers above; they may belong to different providers. These are base rates, before conditional pricing, discounts or additional fees.

Endpoint price comparisonLatest observed snapshot · 4 lowest-priced configurations

InputOutput

Novitanovita/fp8

$0.07$0.27

SiliconFlowsiliconflow/fp8

$0.07$0.28

Amazon Bedrockamazon-bedrock

$0.15$0.60

Alibabaalibaba

$0.2925$1.4625

qwen/qwen3-coder-30b-a3b-instructObserved Oct 4, 2026, 11:31 AM UTC

Input: $0.07
Output: $0.28

Hosting prices through OpenRouter All endpoints and cache rates

Endpoint	Input / 1M	Output / 1M	Cache read / 1M	Cache write / 1M
Alibabaalibaba	$0.2925	$1.4625	—	—
Amazon Bedrockamazon-bedrock	$0.15	$0.60	—	—
Novitanovita/fp8	$0.07	$0.27	—	—
SiliconFlowsiliconflow/fp8	$0.07	$0.28	—	—

Each row is a hosting configuration. Open an offer in Providers for its full pricing conditions and additional fees. Endpoint observations may differ from the routing catalog's base price.

Performance

Review reported latency and output-throughput percentiles from the latest observed rolling window.

Latest 30-minute observation

Time to first token · milliseconds

Latency distributionRolling 30-minute snapshot · 3 endpoints

P50P75P90P99

Alibabaalibaba

1,413.5–13,809.35 ms

Novitanovita/fp8

1,219–23,545 ms

SiliconFlowsiliconflow/fp8

1,812–7,824.6 ms

All latency percentiles 3 endpoints

Endpoint	P50	P75	P90	P99
Alibabaalibaba	1,413.5 ms	2,008 ms	11,701.3 ms	13,809.35 ms
Novitanovita/fp8	1,219 ms	2,225 ms	4,037 ms	23,545 ms
SiliconFlowsiliconflow/fp8	1,812 ms	2,130.5 ms	2,542 ms	7,824.6 ms

Output throughput · tokens per second

Throughput distributionRolling 30-minute snapshot · 3 endpoints

P50P75P90P99

Alibabaalibaba

29–93.33 t/s

Novitanovita/fp8

55–103.27 t/s

SiliconFlowsiliconflow/fp8

20–36 t/s

All throughput percentiles 3 endpoints

Endpoint	P50	P75	P90	P99
Alibabaalibaba	29 t/s	50 t/s	74.9 t/s	93.33 t/s
Novitanovita/fp8	55 t/s	74 t/s	89.7 t/s	103.27 t/s
SiliconFlowsiliconflow/fp8	20 t/s	24 t/s	28 t/s	36 t/s

Percentiles describe the observed request distribution. These are snapshots of a rolling window, not a historical time series.

Uptime

Compare successful request rates across the latest 5-minute, 30-minute, and 24-hour observations.

Successful requests by window

Availability by endpointCurrent overlapping windows · 4 endpoints

Endpoint5 minutes30 minutes24 hours

AlibabaalibabaObserved Oct 4, 2026, 11:31 AM UTC

—

99.86%

Amazon Bedrockamazon-bedrockObserved Oct 4, 2026, 11:31 AM UTC

—

Novitanovita/fp8Observed Oct 4, 2026, 11:31 AM UTC

100%

98.99%

97.57%

SiliconFlowsiliconflow/fp8Observed Oct 4, 2026, 11:31 AM UTC

100%

99.05%

98.6%

Reported by OpenRouter at the time of collection; rate-limited requests are excluded. A dash means no measurement was reported. This is not a live availability check.

Benchmarks

Review attributed evaluation results on each source's original scale.

External evaluations

qwen/qwen3-coder-30b-a3b-instructObserved Oct 4, 2026, 11:31 AM UTC

Design Arena	Category	Elo	Win rate	Rank
models	dataviz	1,091	54.7%	96
models	uicomponent	1,053	54.1%	97
models	website	1,094	57.1%	100

Sources: Artificial Analysis and Design Arena, via OpenRouter. Scores retain their original scales and are not OpenAI Suite ratings.

Detailed evaluations

Evaluation scoresAccuracy or item F1 · higher is better

gpqa diamondQwen: Qwen3 Coder 30B A3B Instruct

52.02%

tau bench verified airlineQwen: Qwen3 Coder 30B A3B Instruct

43.111%

Evaluation	Configuration	Source	Results	Details
models · website	Qwen3 Coder 30B A3B Instructqwen/qwen3-coder-30b-a3b-instruct	Design Arena	1,094 Elo	Details
models · dataviz	Qwen3 Coder 30B A3B Instructqwen/qwen3-coder-30b-a3b-instruct	Design Arena	1,091 Elo	Details
gpqa diamond	Qwen: Qwen3 Coder 30B A3B Instructqwen/qwen3-coder-30b-a3b-instruct	OpenRouter	52.02%	Details
models · uicomponent	Qwen3 Coder 30B A3B Instructqwen/qwen3-coder-30b-a3b-instruct	Design Arena	1,053 Elo	Details
tau bench verified airline	Qwen: Qwen3 Coder 30B A3B Instructqwen/qwen3-coder-30b-a3b-instruct	OpenRouter	43.111%	Details

Sources: Artificial Analysis, Design Arena and OpenRouter. Evaluation configurations and scales differ; results are not interchangeable. Updated Oct 4, 2026.

Apps & session costs

Compare observed median session costs for applications using this model, grouped by conversation length.

30-day sample · through Sep 27, 2026

Application	Model route	Turns	Median cost / session	Details
Hermes Agent	qwen/qwen3-coder-30b-a3b-instruct	2–9 turns	$0.004162	Details
Hermes Agent	qwen/qwen3-coder-30b-a3b-instruct	10–49 turns	$0.050036	Details

Source: OpenRouter, as of Oct 2, 2026. Published under CC BY 4.0. These are observed costs per session, not token prices or a forecast for your workload. Apps are compared separately.

Technical details & API

Explore specifications, supported parameters, reasoning controls, and API identifiers for each OpenRouter route.

Current route specifications

qwen/qwen3-coder-30b-a3b-instructObserved Oct 4, 2026, 11:31 AM UTC

Context: 262,144
Maximum output: 235,929
Tokenizer: Qwen3
Input: text
Output: text
Moderation: No
Knowledge cutoff: 2025-06-30
Added to OpenRouter: 2025-07-31

Base URLhttps://openrouter.ai/api/v1

Model IDqwen/qwen3-coder-30b-a3b-instruct

Version IDqwen/qwen3-coder-30b-a3b-instruct

Weights IDQwen/Qwen3-Coder-30B-A3B-Instruct

Supported parameters

frequency_penalty
logprobs
max_tokens
presence_penalty
repetition_penalty
response_format
seed
stop
structured_outputs
temperature
tool_choice
tools
top_k
top_logprobs
top_p

Frequently asked questions

Answers based on the model metadata and serving offers currently in the catalog.

How much does Qwen3-Coder 30B-A3B Instruct cost?

Listed rates start at $0.00 per million input tokens and $0.00 per million output tokens. Prices vary by provider and configuration; see Pricing for details.

Which providers offer Qwen3-Coder 30B-A3B Instruct?

There are 15 cataloged offers across 12 providers. Compare model IDs, prices and limits in Providers.

What is the context window?

The cataloged model context is 262,144 tokens. Each serving endpoint may apply a different limit.

Does it support tools and structured output?

Tool calling: Yes. Structured output: Not reported. Support can vary by endpoint.

Resources

Hugging Face ↗

Catalog history · 2 recorded revisions

Metadata observations since this model was first cataloged. These are not model release versions.

Sep 14, 2026 · cbaa6f68d2f2
Sep 12, 2026 · cdbe7f675f0f