Qwen3 235B A22B Instruct
DeprecatedQwen3 235B A22B Instruct is Alibaba's language model with a 262K context window and up to 16K output tokens, available from 9 providers, starting at $0.09 / 1M input and $0.55 / 1M output. An instruction-tuned update of the Qwen3 235B A22B MoE model with significant improvements in instruction following, logical reasoning, and general capabilities.
| Benchmarks | |
|---|---|
| Intelligence Index | #276 |
| Math Index | #122 |
| MMLU-Pro | #69 |
| GPQA | #217 |
| HLE | #223 |
| LiveCodeBench | #159 |
| AIME | #42 |
| IFBench | #208 |
| Time to First Token | #25 |
| SciCode | #254 |
| MATH-500 | #22 |
| AIME 2025 | #122 |
| LCR | #308 |
| TerminalBench Hard | #185 |
| TAU2 | #241 |
| Output TPS | #213 |
Capabilities
Input1/5
Text✓
Image·
Audio·
Video·
PDF·
Output1/5
Text✓
Image·
Audio·
Video·
Embedding·
Capabilities3/13
Reasoning·
Adaptive Reasoning·
Function Calling✓
Parallel Function Calling✓
Structured Outputs✓
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching·
Assistant Prefill·
Pricing by Provider
US Dollar ($)
Per 1M tokens
| Provider | Standard | Batch | ||
|---|---|---|---|---|
| Input $ / 1M | Output $ / 1M | Input $ / 1M | Output $ / 1M | |
| $3.00 | $3.00 | — | — | |
| $0.09 | $0.55 | — | — | |
| $0.22 | $0.88 | — | — | |
| $0.22 | $0.88 | $0.11 | $0.44 | |
| $0.855 | $2.56 | — | — | |
| $0.2 | $0.6 | — | — | |
| $0.09 | $0.58 | — | — | |
| $0.264 | $1.06 | — | — | |
| $0.75 | $2.25 | — | — | |
Cost Calculator
US Dollar ($)
Preset:
Cheapest Instances to Run It
Cloud GPU instances that can host Qwen3 235B A22B Instruct, ranked by cheapest on-demand price. The model needs about 564 GB of GPU memory at FP16 precision (estimated from its parameter count), so treat the fit as guidance rather than a guarantee.
All clouds
FP16 (full precision)
US Dollar ($)
Instance | Cloud | GPU | VRAM | Price | Cheapest region | |
|---|---|---|---|---|---|---|
| p4de.24xlarge |  AWS | 8× A100 | 640 GB | $27.45/hr | ||
| Standard_ND96amsr_A100_v4 |  Azure | 8× NVIDIA A100 (80GB) | 640 GB | $32.77/hr | ||
| g7e.48xlarge | AWS | 8× RTX PRO Server 6000 | 768 GB | $33.14/hr | ||
Versions
| Version | Released | Context | Input / 1M | Output / 1M | Status |
|---|---|---|---|---|---|
| MiroThinker 1.7 | — | — | — | — | Available |
| Qwen3 30B A3B Thinking | 262K | $0.900 | $0.900 | Deprecated | |
| Qwen3 Coder 30B A3B Instruct | 262K | $0.070 | $0.270 | Deprecated | |
| Qwen3 30B A3B Instruct | 262K | $0.100 | $0.300 | Deprecated | |
| Qwen3 235B A22B Thinking | 262K | $0.220 | $0.880 | Deprecated | |
| Qwen3 Coder | 262K | $0.220 | $1.00 | Deprecated | |
| Qwen3 235B A22B | 262K | $0.090 | $0.540 | Deprecated | |
| Qwen3 30B A3B | 131K | $0.051 | $0.200 | Deprecated | |
| Qwen3 235B A22B Instruct | — | 262K | $0.090 | $0.550 | Current |
| Qwen3 Coder 480B A35B Instruct | — | 262K | $0.220 | $1.55 | Deprecated |
| KwaiPilot KAT Coder | — | 262K | $0.900 | $0.900 | Available |