Qwen3 30B A3B Instruct 2507
Qwen instruction model for multilingual chat, reasoning, and tool use
Shogo Agent Index
UnrankedNot ranked yet: 0 agentic benchmarks measured, and we need at least 2 plus an overall capability score. It will be ranked as results are published.
Agentic coding
17
estimated
Long-horizon autonomy
28
estimated
Knowledge-work agents
16
estimated
Computer use
6
estimated
Security & context learning
16
estimated
Agentic benchmarks
| Benchmark | Score | Normalised | Index weight | Harness / source |
|---|---|---|---|---|
| SWE-bench Verified | est. | 8% | predicted 50 | |
| Terminal-Bench 2.0 | est. | 8% | predicted 8 | |
| DeepSWE | est. | 6% | predicted 9 | |
| FrontierSWE | est. | 5% | predicted 1 | |
| CursorBench | est. | 4% | predicted 17 | |
| GSO-Bench | est. | 4% | predicted 2 | |
| METR Time Horizons | est. | 10% | predicted 50 | |
| Vending-Bench 2 | est. | 6% | predicted 2 | |
| PostTrainBench | est. | 4% | predicted 11 | |
| APEX-Agents | est. | 8% | predicted 11 | |
| GDPval | est. | 7% | predicted 17 | |
| Remote Labor Index | est. | 5% | predicted 0 | |
| DeepResearch Bench | est. | 5% | predicted 41 | |
| OSWorld 2.0 | est. | 6% | predicted 1 | |
| OSWorld | est. | 4% | predicted 15 | |
| Cybench | est. | 5% | predicted 20 | |
| CL-bench | est. | 5% | predicted 12 |
Other benchmarks
| Benchmark | Category | Score | Source |
|---|---|---|---|
| DTBench | Knowledge | 67.2% | source |
| GPQA Diamond | Reasoning | 55.6% | source |
| LMCA | Knowledge | 22.4% | source |
| OTIS Mock AIME | Math | 62.2% | source |
API hosts
Live prices and uptime for every host serving Qwen3 30B A3B Instruct 2507 via OpenRouter.
| Host | Input | Output | Cache read | Agent $/M | Uptime 24h | Quant |
|---|---|---|---|---|---|---|
| DekaLLM | $0.090 | $0.30 | — | $0.10 | 99.0% | — |
| SiliconFlow | $0.090 | $0.30 | — | $0.10 | 99.2% | fp8 |
| Nebius | $0.10 | $0.30 | $0.10 | $0.11 | 96.7% | fp8 |