UpstageAuto-detectedSolar Pro 3
Best configuration ranks #412 of 610 on the BenchLeader Index at 44.6 ±11.0. Last measured 27 Jan 2026stale: no new result in six months. Released 27 Jan 2026.
- Blended price
- $0.262/M
- $0.150 in · $0.600 out
- Output speed
- 147 tok/s
- First answer
- 16 s
- first token 0.85 s
- Context
- 131k
- Overall index45
- Coding23
- Agents & tools40
- Knowledge39
- Instruction following69
- Long context42
- Composite39
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | default | Source |
|---|---|---|
| GPQA Diamond (AA)not in index | 72.4%#270 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 10.3%#252 | Artificial Analysis |
Coding
| Benchmark | default | Source |
|---|---|---|
| SciCode | 24.6%#151 | SciCode |
| SciCode (AA)not in index | 25.5%#152 | Artificial Analysis |
Agents & tools
| Benchmark | default | Source |
|---|---|---|
| Terminal-Bench Hard | 7.6%#232 | Artificial Analysis |
| τ²-Bench Telecom (AA)not in index | 86.3%#87 | Artificial Analysis |
Knowledge
| Benchmark | default | Source |
|---|---|---|
| AA-Omniscience | -53.4#345 | Artificial Analysis |
Instruction following
| Benchmark | default | Source |
|---|---|---|
| IFBench | 71.2%#59 | Artificial Analysis |
Long context
| Benchmark | default | Source |
|---|---|---|
| AA-LCR | 32.3%#312 | Artificial Analysis |
Composite
| Benchmark | default | Source |
|---|---|---|
| AA Intelligence Index | 7.8#378 | Artificial Analysis |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Upstage | 11 tok/s | 0.85 s | $0.150 | $0.600 | 131k | – |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0002 | – | 18.3 s |
| Summarise a 30-page report | 12,000 / 600 | $0.0022 | – | 20.3 s |
| Code edit | 6,000 / 1,500 | $0.0018 | – | 26.5 s |
| Agentic coding session | 60,000 / 4,000 | $0.011 | – | 43.5 s |
| Structured extraction | 2,000 / 200 | $0.0004 | – | 17.6 s |
See also
Data as of 9 Sept 2026. Compare with another model.