Solar-Open2-250B: benchmark scores, price and ranking
Solar-Open2-250B from upstage ranks #49 of 743 LLMs on the LLMs Tiger aggregate benchmark score (52.9), based on 2 benchmark results captured up to 2026-10-13.
- Organisation
- upstage
- Type
- Open-weight / local model
- Open weights
- Yes
- Aggregate rank
- #49 of 743 (score 52.9)
Across the 2 benchmarks where at least ten models are scored, Solar-Open2-250B is in the top 10% on 1 and in the top half on 2.
Solar-Open2-250B benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| MMLU-Pro | 86.2% | #10 of 136 | HuggingFace MMLU-Pro leaderboard | 2026-10-13 |
| GPQA Diamond | 86.3% | #67 of 275 | HuggingFace GPQA Diamond leaderboard | 2026-10-13 |
About Solar-Open2-250B
Solar-Open2-250B is an open-weight model that can run on your own hardware from upstage. No API price is listed for it yet. We hold 2 benchmark results for it across 1 area, and 1 is in the top tenth of its board.
Where Solar-Open2-250B stands
On our aggregate score, Solar-Open2-250B is #49 of 743 ranked models, in the top fifty. Its score is 52.9; the scale is centred on 50, so it sits 2.9 points above the typical model. That beats 93% of the ranked field.
Solar-Open2-250B, benchmark by benchmark
On MMLU-Pro, Solar-Open2-250B scores 86.2% and ranks #10 of 136. Only 9 models score higher, led by Darwin-180B-RSI at 88.1%, which is 1.9 points ahead. It covers 57-subject knowledge + reasoning. The board median is 55.3%, so Solar-Open2-250B is 30.9 points above it. Source: HuggingFace MMLU-Pro leaderboard, captured 2026-10-13.
On GPQA Diamond, Solar-Open2-250B reaches 86.3%. It ranks #67 of 275, ahead of 76% of the field, and 9.5 points behind GPT-6 Astra (max) at 95.8%. The results either side are Claude Opus 4.7 (max) (86.4%) and GLM-5.1 (86.2%). It covers expert-level science questions. The board median is 71.7%, so Solar-Open2-250B is 14.6 points above it. Source: HuggingFace GPQA Diamond leaderboard, captured 2026-10-13.
Strengths and weaknesses
Reasoning, maths and science: strong, ahead of 84% of the field on average across 2 benchmarks. Its best board here is MMLU-Pro (#10 of 136) and its weakest is GPQA Diamond (#67 of 275).
Reading these numbers
The 2 scores on this page come from 2 sources: HuggingFace GPQA Diamond leaderboard (1), HuggingFace MMLU-Pro leaderboard (1). No official leaderboard supplied any of them, so treat them as third-party measurements. With 2 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #46 Qwen3.5-122B-A10B (53.0)
- #47 Claude Mythos Preview (Early) (53.0)
- #48 Gemini 2.5 Pro Preview 05-06 (53.0)
- #50 Kimi K2.5 (Fireworks) (52.8)
- #51 CodeAct v2.1 (claude-3-5-sonnet-20241022) (52.8)
- #52 claude-3.5-sonnet-20240620 (52.7)
Compare it on the live leaderboard → · How the aggregate score works