Examenos

Official vendor benches · independent evaluators · OpenRouter pricing

Model profilesv d0b1ff3

Solar Mini 4

Explore Solar Mini 4 from Upstage: published specifications, source-linked vendor benchmarks, independent evaluator coverage and recorded pricing when available.

Compare Solar Mini 4 with other models →Compare subscription plansExplore data coverage

Pricing · OpenRouter

Dated cached OpenRouter rates in USD per 1M tokens. Open the dashboard for live enhancements. Per-metric endpoint minima can refer to different providers; they are not a guaranteed combined rate from one endpoint.

Recorded pricing tiers
TierInput / 1MOutput / 1MCached input / 1MCache write / 1MDate & source
Default$0.05$0.2$0.005—2026-11-01 · OpenRouter source
Recorded pricing notes

OpenRouter list rate captured 1 November 2026, pending the next min-healthy refresh. No contributor sibling. OpenRouter canonical upstage/solar-mini4-20260922; Upstage standard pricing is $0.10/$0.40/$0.01, with a launch discount noted on its blog through 22 October.

Official / vendor benchmarks

Default headline records. Own-vendor, peer-vendor and third-party provenance remain visible in evidence; configurations may differ.

Official / vendor headline scores; expand evidence for every record
Benchmark / evaluatorHeadline scoreEvidence
AA Intelligence Index v4.3.2 (vendor-cited)24.1unknown effort
All 1 recorded result & sources

24.1 · raw 24.1 index

Headline · unknown effort · Own vendor

Source/record date: 2026-10-01

Upstage launch article; vendor-reported, not a direct Artificial Analysis observation. Evaluation date unpublished.

https://www.upstage.ai/blog/en/solar-mini-4
AutomationBench-AA (vendor-reported)22.3%unknown effort
All 1 recorded result & sources

22.3% · raw 22.3 %

Headline · unknown effort · Own vendor

Source/record date: 2026-10-01

Upstage launch article; vendor-reported, not a direct Artificial Analysis observation. Evaluation date unpublished.

https://www.upstage.ai/blog/en/solar-mini-4
τ³-bench Banking47.2%unknown effort
All 1 recorded result & sources

47.2% · raw 47.2 %

Headline · unknown effort · Own vendor

Source/record date: 2026-10-01

Upstage reports 47.2 in a separate τ³-Banking evaluation; harness and unit are not stated, so the source figure is treated as a percent pass rate as on Solar Pro 4. Evaluation date unpublished.

https://www.upstage.ai/blog/en/solar-mini-4
AA-LCR83.3%unknown effort
All 1 recorded result & sources

83.3% · raw 83.3 %

Headline · unknown effort · Own vendor

Source/record date: 2026-10-01

Upstage launch article; vendor-reported, not a direct Artificial Analysis observation. Evaluation date unpublished.

https://www.upstage.ai/blog/en/solar-mini-4
SciCode47.6%unknown effort
All 1 recorded result & sources

47.6% · raw 47.6 %

Headline · unknown effort · Own vendor

Source/record date: 2026-10-01

Upstage launch article; vendor-reported, not a direct Artificial Analysis observation. Evaluation date unpublished.

https://www.upstage.ai/blog/en/solar-mini-4
Humanity's Last Exam (tool setting unspecified)25.8%unknown effort
All 1 recorded result & sources

25.8% · raw 25.8 %

Headline · unknown effort · Own vendor

Source/record date: 2026-10-01

Upstage launch article; vendor-reported, not a direct Artificial Analysis observation. Evaluation date unpublished. Tool setting unspecified.

https://www.upstage.ai/blog/en/solar-mini-4

Independent evaluators

Evaluator harnesses are distinct from vendor measurements. Missing coverage is not a failed test.

Independent evaluator headline scores; expand evidence for every record
Benchmark / evaluatorHeadline scoreEvidence
DataAnalyticsBench · questions correct /43 · Plotly DataAnalyticsBench21 correctunknown effort
All 1 recorded result & sources

21 correct · raw 21 correct

Headline · unknown effort · Independent evaluator

Source/record date: 2026-11-01

Harness: Plotly Studio Embedded, one run on 43 private English data-analytics questions in a synthetic wind-energy warehouse. Claude Sonnet 5 judges against Plotly answer key. One or two questions can be noise; no error bars. Exact checkpoint hash, run date and effort unpublished. Company-operated analytics benchmark, distinct from provider results. Plotly names this result in its 31 October update text.

https://plotly.com/blog/introducing-dataanalyticsbench/

Read how we select and source scores or the comparison guide.