Capability
Unranked
field median 56.2
Not eligible for a public rank
Keep up with the models you depend on. Follow price changes, retirements, and API updates.Follow the models you depend on.
Follow model changesReleased Jul 22, 2026 — see all recent releases
Data as of September 10, 2026 · How the score is built
Share or export
Published rows are visible, but no category has enough eligible evidence for a comparative rank.
14 published rows leave some tracked benchmark slots empty. No comparable first-party API token rate is published.
Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.
Capability
Unranked
field median 56.2
Not eligible for a public rank
Price
Self-hosted; infrastructure cost varies
input median $1
No comparable first-party hosted token rate
Speed
Not measured
field median 86.5 tok/s
Time to first token not measured
Context
1Mtokens
field median 256,000
Maximum output length is tracked separately
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Published weights are available, but BenchLM does not yet have a sourced parameter and VRAM profile for this exact model. Hardware cost estimates stay unavailable until that sizing record is complete.
Estimate VRAM from known parametersScores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticWeight 22%3 benchmarksVerified | Score pending | Not ranked | Not available | 22% | 3 benchmarks | Verified |
| CodingRank Not rankedWeight 20%2 benchmarksVerified | 48.4 | Not ranked | Not available | 20% | 2 benchmarks | Verified |
| ReasoningWeight 17%0 benchmarksNot measured | Not measured | Not ranked | Not available | 17% | 0 benchmarks | Not measured |
| KnowledgeRank Not rankedWeight 12%3 benchmarksVerified | 71.7 | Not ranked | Not available | 12% | 3 benchmarks | Verified |
| MathRank Not rankedWeight 5%2 benchmarksVerified | 79.1 | Not ranked | Not available | 5% | 2 benchmarks | Verified |
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | Not ranked | Not available | 7% | 0 benchmarks | Not measured |
| MultimodalWeight 12%0 benchmarksNot measured | Not measured | Not ranked | Not available | 12% | 0 benchmarks | Not measured |
| Inst. FollowingRank Not rankedWeight 5%1 benchmarkVerified | 85.6 | Not ranked | Not available | 5% | 1 benchmark | Verified |
Coding opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| SWE-bench VerifiedSoftware Engineering Benchmark Verified | Score70.4% | Versus best verified row Best verified: Claude Opus 5 · 96% | Gap25.6 behind | WeightWeighted 10% | Provider exact |
| LiveCodeBench v6 | Score92.4% | Versus best verified row Best verified: Sakana Fugu-Ultra · 93.2% | Gap0.8 behind | WeightDisplay only | Provider exact |
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| terminalBenchHard | Score28.3% | Versus best verified row Best verified: Solar Open 2 · 28.3% | GapBest verified | WeightDisplay only | Provider exact |
| APEX-Agents | Score16.6% | Versus best verified row Best verified: Grok 4.6 · 57.5% | Gap40.9 behind | WeightDisplay only | Provider exact |
| MCP Atlas | Score58.2% | Versus best verified row Best verified: Muse Spark 1.1 · 88.1% | Gap29.9 behind | WeightDisplay only | Provider exact |
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| MMLU-ProMassive Multitask Language Understanding Professional | Score86.2% | Versus best verified row Best verified: Qwen3.7 Max · 89.6% | Gap3.4 behind | WeightWeighted 20% | Provider exact |
| HLE w/o toolsHumanity's Last Exam without tools | Score28.8% | Versus best verified row Best verified: Claude Fable 5.1 · 60.9% | Gap32.1 behind | WeightWeighted 10% | Provider exact |
| GPQA-DGPQA Diamond | Score86.3% | Versus best verified row Best verified: GPT-6 Astra · 96.0% | Gap9.7 behind | WeightDisplay only | Provider exact |
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| HMMT Feb 2026Harvard-MIT Mathematics Tournament February 2026 | Score93.9% | Versus best verified row Best verified: Qwen3.7 Max · 97.1% | Gap3.2 behind | WeightWeighted 25% | Provider exact |
| AIME26AIME 2026 | Score95.7% | Versus best verified row Best verified: GLM-5.2 · 99.2% | Gap3.5 behind | WeightWeighted 25% | Provider exact |
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| IFBenchInstruction Following Benchmark | Score80% | Versus best verified row Best verified: MAI-Thinking-1 · 85% | Gap5 behind | WeightWeighted 70% | Provider exact |
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| KMMLU-Pro | Score78.4% | Versus best verified row Best verified: A.X K2 · 80.5% | Gap2.1 behind | WeightDisplay only | Provider exact |
| CLIcKCultural and Linguistic Intelligence in Korean | Score90.7% | Versus best verified row Best verified: A.X K2 · 91.6% | Gap0.9 behind | WeightDisplay only | Provider exact |
| HRM8KHAE-RAE Math 8K | Score92.2% | Versus best verified row Best verified: Solar Open 2 · 92.2% | GapBest verified | WeightDisplay only | Provider exact |
The sequence follows explicit supersedes links. A successor's displayed score stays at least 0.1 points above its predecessor; raw benchmark rows do not move. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Jul 22, 2026 · you are here
Solar Open 2Not publicly ranked · Price not listed
Base entry
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
We track Solar Open 2, but the public leaderboard excludes this profile until enough non-generated benchmark coverage is available. Only published rows appear above.
Solar Open 2 is a open weight model with a 1M context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
Upstage publishes the 250B-total, 15B-active hybrid-attention MoE checkpoint on Hugging Face under the Upstage Solar License (an Apache 2.0 derivative that permits commercial use and derivatives) with a 1M-token context window and Korean, English, and Japanese support. Recommended serving is vLLM on 4-8 H200 GPUs.
14 of 434 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Upstage · Model release
Radar confirmed these at the source. Use Solar Open 2 in your work? Explore Radar to follow supported changes and choose your alerts.
Solar Open 2 has 14 source-displayable benchmark rows, but it does not qualify for a public overall rank. The available rows remain visible by category without being converted into a site-wide score. Missing evidence stays blank instead of being estimated from an earlier model.
Solar Open 2 has source-displayable benchmark coverage for knowledge and understanding, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Solar Open 2 has source-displayable benchmark coverage for coding and programming, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Solar Open 2 has source-displayable benchmark coverage for mathematics, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Solar Open 2 has source-displayable benchmark coverage for agentic tool use and computer tasks, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Solar Open 2 has source-displayable benchmark coverage for instruction following, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Solar Open 2 has 3 source-displayable Korean benchmark scores on this profile. Those rows remain separate from the global overall score because regional evaluations use a different comparison set. Use the dedicated Korean tables for side-by-side positions and inspect each source before applying the result elsewhere.
Solar Open 2 is an open-weight model from Upstage. Its weights can be downloaded for local or hosted deployment, subject to the published license. Open weight does not automatically mean open source: training data and training code may remain private, and commercial restrictions can still apply.
No. Solar Open 2 currently has 15 source-displayable rows across 434 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
Solar Open 2 has a documented context window of 1M. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.
Related resources
Last updated September 10, 2026. Runtime fields remain blank until a sourced snapshot exists.
Get one weekly email when material rank, price, availability, or benchmark evidence changes are worth revisiting.
Read a sample issueJoin 2,000+ readers.