Qwen: Qwen3 Coder 30B A3B Instruct

codeapache-2.030B / 3B

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.

Context
262K
Max output
236K
Input price
$0.07/1M tokens
Output price
$0.28/1M tokens

Capabilities

Reasoning & Interaction
Reasoning
Tool Calling
Output Control
Structured Output
JSON Mode
Infrastructure
Streaming
Batch Processing
Fine-tuning

Details

Model IDqwen/qwen3-coder-30b-a3b-instruct
Provider OpenRouter
Creator Qwen
Familyqwen3
Licenseapache-2.0
Status—
Typecode
Input modalitiestext
Output modalitiestext
Open weightYes
Parameters30B / 3B
Architecture—
Context window262K tokens
Max input—
Max output236K tokens
Reasoning tokens—
Knowledge cutoff—
Release date
Sourceofficial

Tools

Function Callingfunction_calling
Call external functions and APIs

Pricing

Input
$0.07
Output
$0.28
Cache write
—
Cache read
—
Batch in
—
Batch out
—

Family Comparison: qwen3

ModelContextMax outPricing
Qwen: Qwen3 4B (free)41K——
Qwen: Qwen3 Coder 480B A35B (free)1.0M262K—
Qwen: Qwen3 Next 80B A3B Instruct (free)262K——
Qwen: Qwen3 Coder 30B A3B Instruct262K236K$0.07$0.28
Qwen: Qwen3 32B131K16K$0.08$0.28
Qwen: Qwen3 235B A22B Instruct 2507262K16K$0.09$0.55
Qwen: Qwen3 Next 80B A3B Instruct262K16K$0.09$1.1
Qwen: Qwen3 30B A3B Instruct 2507262K236K$0.1$0.3
Qwen: Qwen3 VL 32B Instruct131K33K$0.1$0.42
Qwen: Qwen3 8B131K8K$0.12$0.46
Qwen: Qwen3 14B41K16K$0.12$0.24
Qwen: Qwen3 30B A3B41K16K$0.12$0.5
Qwen: Qwen3 Coder Next262K236K$0.12$0.8
Qwen: Qwen3 Next 80B A3B Thinking262K236K$0.15$1.2
Qwen: Qwen3 VL 30B A3B Instruct262K16K$0.15$0.6
Qwen: Qwen3 VL 8B Thinking131K33K$0.18$2.1
Qwen: Qwen3 Coder Flash1M66K$0.2$0.98
Qwen: Qwen3 30B A3B Thinking 250782K33K$0.2$2.4
Qwen: Qwen3 VL 235B A22B Instruct262K33K$0.21$1.9
Qwen: Qwen3 VL 8B Instruct262K236K$0.25$0.75
Qwen: Qwen3 VL 30B A3B Thinking262K236K$0.29$1
Qwen: Qwen3 Coder 480B A35B262K66K$0.3$1
Qwen: Qwen3 VL 235B A22B Thinking131K33K$0.4$4
Qwen: Qwen3 235B A22B Thinking 2507128K16K$0.45$3.5
Qwen: Qwen3 Coder Plus1M66K$0.65$3.25
Qwen: Qwen3 Max Thinking262K66K$0.78$3.9
Qwen: Qwen3 Max262K66K$0.78$3.9

API

GET/v1/models/openrouter/qwen/qwen3-coder-30b-a3b-instruct

Qwen: Qwen3 Coder 30B A3B Instruct

qwen/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.

Changes · 10 entries
qwen/qwen3-coder-30b-a3b-instructupdateca7a400Nov 2, 2026, 03:02 AM
context_window160000→262144
max_output_tokens32768→235929
pricingoutput$0.27→$0.28
qwen/qwen3-coder-30b-a3b-instructupdatee3d7dedAug 31, 2026, 04:13 AM
max_output_tokens262144→235929
qwen/qwen3-coder-30b-a3b-instructupdate3cddff5Aug 13, 2026, 02:54 AM
max_output_tokens32768→262144
pricingoutput$0.27→$0.28
qwen/qwen3-coder-30b-a3b-instructupdate19c9fe3Aug 5, 2026, 03:04 PM
max_output_tokens262144→32768
pricingoutput$0.28→$0.27
qwen/qwen3-coder-30b-a3b-instructupdate86db01cAug 2, 2026, 08:20 AM
max_output_tokens32768→262144
pricingoutput$0.27→$0.28
qwen/qwen3-coder-30b-a3b-instructupdatebf78dcfJul 25, 2026, 05:49 AM
context_window160000→262144
qwen/qwen3-coder-30b-a3b-instructupdate1cd659eMar 24, 2026, 04:19 AM
capabilitiesjson modeyes
qwen/qwen3-coder-30b-a3b-instructupdate2b100acMar 23, 2026, 05:30 AM
parameters30
active_parameters3
release_date2025-07-31
taglineQwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture...
tools["function_calling"]
qwen/qwen3-coder-30b-a3b-instructupdate7a4f1f6Mar 22, 2026, 05:43 AM
model_typecode
qwen/qwen3-coder-30b-a3b-instructcreate7f5c66fMar 21, 2026, 05:16 AM