Upstage
availableShows if the model has enough results for an index.Solar Pro 4
Solar Pro 4 is a reasoning model from Upstage in the Solar family. 17 benchmarks count toward its score, in 5 categories.
IndexOverall score. 50 is the middle.53.7 ±5.9
CoverageShare of the index weight with results.80%
SpeedOutput tokens per second.38/s
Input / 1MUS dollars per 1M input tokens.$0.09
Output / 1MUS dollars per 1M output tokens.$0.36
ContextMaximum tokens in one request.524K
EloLMArena rating and rank.1377 (#161)
50 is the middle of the board. The range shows the doubt in the index.
2,420 votes. Elo shows what people prefer. It does not change the score.
CapabilitiesScore per category. 50 is the middle.
50 is the middleResults
17 counted| BenchmarkThe test name. | CategoryThe capability that the test measures. | ResultThe score from the publisher. | IndexThis result on the index scale. | RunThe settings of the run. | DateDate of the result. | Published byThe source of the result. |
|---|---|---|---|---|---|---|
| AIME 2026 | Math | 95.3% | 54.4 | — | — | Qwen |
| Artificial Analysis GPQA Diamond | Knowledge | 89.1% | 59.9 | — | — | Artificial Analysis |
| GPQA Diamond | Knowledge | 89.0% | 60.4 | — | — | David Rein et al. |
| LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code | Coding | 87.8% | 64.1 | — | — | Naman Jain et al. |
| Massive Multitask Language Understanding Professional | Knowledge | 86.3% | 56.5 | — | — | Yubo Wang et al. |
| Artificial Analysis Long Context Reasoning | Reasoning | 71.0% | 57.4 | — | — | Artificial Analysis |
| Software Engineering Benchmark Verified | Coding | 70.6% | 54.2 | — | — | Carlos E. Jimenez et al. |
| MCP Atlas | Agentic | 61.4% | 54.4 | — | — | OpenAI |
| Terminal-Bench 2.1 (provider run) | Agentic | 57.0% | 57.7 | — | — | DeepSeek-AI |
| Terminal-Bench 2.1 (provider run) | Agentic | 57.0% | 57.7 | — | — | DeepSeek-AI |
| BrowseComp | Agentic | 49.2% | 41.3 | — | — | OpenAI |
| Artificial Analysis SciCode | Coding | 44.6% | 54.8 | — | — | Artificial Analysis |
| Artificial Analysis Humanity's Last Exam | Knowledge | 29.2% | 56.2 | — | — | Artificial Analysis |
| Artificial Analysis Intelligence Index | Knowledge | 28.1% | 57.6 | — | — | Artificial Analysis |
| Artificial Analysis Omniscience Accuracy | Knowledge | 18.9% | 37.3 | — | — | Artificial Analysis |
| APEX-Agents | Agentic | 18.7% | 55.1 | — | — | Moonshot AI / APEX-Agents benchmark authors |
| Critical Physics Tasks | Reasoning | 5.4% | 49.7 | — | — | Artificial Analysis |
| Agent Arena steerability | Agentic | -10.1 | 56.0 | — | 15 Sept 2026 | LMArena |
| Agent Arena task outcome | Agentic | -20.1 | 44.7 | — | 15 Sept 2026 | LMArena |
| Agent Arena command recovery | Agentic | -24.2 | 40.1 | — | 15 Sept 2026 | LMArena |
17 benchmarks count, from 20 of 20 results. A grey row does not count. Too few models took that benchmark.