Upstageproprietary
Released: 2024-11-20Solar Pro
Upstage 22B high-throughput foundation model designed for single-GPU enterprise hosting.
Intelligence Index
78.0/ 100
Calibrated multi-domain compositeThroughput Speed
90tok/sec
Output streaming throughputTime To First Token
290ms
Initial chunk server latencyLiveBench Contam-Free
68.0%
September 2026 suite scoreSourced Benchmark Evaluations (23)
Verified performance across authoritative benchmarks with provenance tracking.| Benchmark | Status | Score | Trust | Date | Source Type | Provenance |
|---|---|---|---|---|---|---|
| LiveBench (Continuous) | Active | 68% | 94 | 2025-02-01 | independent | Source ↗ |
| StructFlowBench | Nearing Saturation | 61% | 75 | 2024-12-08 | independent | Source ↗ |
| FireBench | Active | 60.3% | 95 | 2024-12-06 | independent | Source ↗ |
| SimpleQA | Active | 40.6% | 90 | 2024-12-05 | independent | Source ↗ |
| OpenAI-MRCR | Active | 65.5% | 80 | 2024-12-05 | independent | Source ↗ |
| ComplexBench | Nearing Saturation | 58.8% | 80 | 2024-12-04 | independent | Source ↗ |
| LongBench-Pro | Active | 64% | 80 | 2024-12-04 | independent | Source ↗ |
| ZeroSCROLLS | Nearing Saturation | 62.4% | 78 | 2024-12-03 | independent | Source ↗ |
| InfiniteBench | Nearing Saturation | 66.3% | 65 | 2024-12-02 | independent | Source ↗ |
| AgentIF | Nearing Saturation | 48.8% | 80 | 2024-12-02 | independent | Source ↗ |
| MMLU-Pro | Nearing Saturation | 71.8% | 75 | 2024-12-01 | independent | Source ↗ |
| SCROLLS | Nearing Saturation | 62.4% | 65 | 2024-12-01 | independent | Source ↗ |
| RULER | Nearing Saturation | 71.8% | 84 | 2024-11-30 | independent | Source ↗ |
| MMLU-Redux | Saturated | 76.4% | 25 | 2024-11-29 | independent | Source ↗ |
| QuALITY | Nearing Saturation | 68.6% | 78 | 2024-11-29 | independent | Source ↗ |
| LongBench | Nearing Saturation | 67.1% | 65 | 2024-11-28 | independent | Source ↗ |
| QASPER | Nearing Saturation | 64.7% | 65 | 2024-11-27 | independent | Source ↗ |
| TruthfulQA | Saturated | 68.6% | 25 | 2024-11-27 | independent | Source ↗ |
| MT-Bench | Nearing Saturation | 7.4% | 75 | 2024-11-25 | independent | Source ↗ |
| IFEval | Deprecated | 70.3% | 25 | 2024-11-24 | independent | Source ↗ |
| GPQA | Nearing Saturation | 68.6% | 78 | 2024-11-24 | independent | Source ↗ |
| Needle-in-a-Haystack | Saturated | 95.6% | 44 | 2024-11-23 | independent | Source ↗ |
| MMLU | Saturated | 79.6% | 25 | 2024-11-21 | independent | Source ↗ |
OpenAI Compatible API ExecutionAPI Ready
from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(
model="solar-pro",
messages=[{"role": "user", "content": "Evaluate multi-step logic problem"}],
temperature=0.2
)
print(response.choices[0].message.content)Evaluation & Sourcing Notes
Scores listed for Solar Pro represent verified evaluations extracted from official research papers, independent evaluation suites (HELM, LMSYS, OpenCompass, LiveBench), and verified audit reports.
All benchmarks marked as saturated or deprecated reflect historical performance where the benchmark no longer provides active discriminative power.