Today's price · per 1M tokens
Input
$0.150
per 1M tokens
Output
$0.600
per 1M tokens
Blended
$0.262
blended $/1M — 3:1 weighted input:output
Cached input
$0.015
10% of input — prompt caching
How this price is scoped: Upstage's own pricing page list price, per 1M tokens (input $0.15, cached input $0.015, output $0.60), which OpenRouter matches exactly. Upstage states that all prices shown are exclusive of 10% VAT, so a VAT-liable invoice runs 10% above the tracked rate. Third-party catalogue models.dev disagrees, listing this SKU at a flat $0.25 in and out; the vendor's own page is the source of record here. Solar Pro 2 is listed by Upstage at the same $0.15/$0.60. Upstage published Solar Pro 4 on 2026-08-11 (tracked as upstage-solar-pro-4) and Solar Pro 3 remains listed at this price alongside it; the earlier note here, that no 'Solar Pro 4' appeared on Upstage's own pricing page as of 2026-08-07, was true when written and is superseded by that launch. Upstage's pricing page moved to /pricing/api — the bare /pricing URL cited here previously now 404s.
Price receipt
No confirmed receipt. We hold dated captures of Upstage's pricing page, but none of them tied Solar Pro 3 to the price above on that page, so we do not claim this number is sourced. Check the provider's page directly before relying on it.
Overview
Upstage's current flagship Solar LLM — a 102B-parameter mixture-of-experts model activating 12B parameters per token, tuned for reasoning, instruction following and bilingual Korean/English work, and updated in March 2026 for agentic use. $0.15/$0.60 per 1M tokens with cached input at $0.015. Upstage lists all prices exclusive of 10% VAT.
Capabilities
struck through = not supportedBenchmark performance
accuracy % · higher is betterWe hold no per-benchmark accuracy scores for this model yet, so it has no accuracy average. That is a gap in our coverage, not a sign the model is untested — independent evaluators often publish a composite index for a new model long before releasing its per-benchmark numbers. The percentile above is its standing across the independent composites below.
- AA Intelligence Index: 7.8 (Artificial Analysis composite across reasoning, knowledge and coding evals)
- AA-Omniscience Index: -53.4 (−100–100)
- Ranks #161 of 181 comparably-measured models by percentile score, across 2 independent measurements
- Ranks #85 of 181 comparably-tested models by normalized performance per dollar
Source: Artificial Analysis · updated · See full rankings →
Specifications
- Provider
- Upstage
- Context window
- 131K tokens
- Modality
- text
- Parameters
- 102B (12B active)
- Open source
- No — proprietary
- Released
- Status
- Current
- Last updated
- Tags
Availability verified: — listed on Upstage's own page