Upstage Solar API Pricing (November 2026)
Upstage Solar API prices run from $0.10 to $1.20 per million tokens across 2 current models. The cheapest is Solar Mini 4 at $0.10 input and $0.40 output per million tokens. The flagship, Solar Pro 4, costs $0.30 input and $1.20 output per million tokens. Every rate below was read on Upstage's own pricing page, on the date shown in its row.
| Model | Input /M | Cache read /M | Output /M | Batch in /M | Batch out /M | Context | Status | Checked |
|---|---|---|---|---|---|---|---|---|
Solar Pro 4 solar-pro4 | $0.30 | $0.06−80% | $1.20 | — | — | 512K | Flagship | 2026-10-23 |
Solar Mini 4 solar-mini4 | $0.10 | $0.01−90% | $0.40 | — | — | 512K | Current | 2026-10-23 |
US dollars per million tokens, list rates on Upstage's own API. Cache read is the price of a cached input token. Batch is the asynchronous API, where the vendor offers one. The date in each row is when we last read it on the vendor's pricing page; it links to that page.
Upstage Solar API monthly cost estimates
What light use (10M input + 2M output tokens a month), medium use (100M input + 20M output tokens a month), heavy use (1B input + 200M output tokens a month) costs on each current model, at list price with no caching and no batch discount.
| Model | Input / output per 1M | Light · 10M + 2M | Medium · 100M + 20M | Heavy · 1B + 200M |
|---|---|---|---|---|
| Solar Mini 4 | $0.10 / $0.40 | $1.80 | $18.00 | $180 |
| Solar Pro 4 | $0.30 / $1.20 | $5.40 | $54.00 | $540 |
Prompt caching and the Batch API cut these figures; the next sections show by how much. For your own token counts, use the LLM API cost calculator.
Upstage Solar API price history
Upstage Solar API price changes and releases
Company milestones and the Solar line. Sourced from Upstage's own newsroom, its pricing and docs pages, and Wikipedia.
Retired Upstage Solar models and their replacements
2 Upstage Solar models no longer answer on Upstage's API. The price is the last one Upstage listed; the replacement is the current end of the vendor's own succession chain.
| Model | Status | Last price in / out per 1M | Use instead |
|---|---|---|---|
| Solar Pro 3 | Retired Oct 30, 2026 | $0.15 / $0.60 | Solar Pro 4 |
| Solar Pro 2 | Retired Oct 30, 2026 | $0.15 / $0.60 | Solar Pro 4 |
console.upstage.ai — API keys, docs and playground. The endpoint is OpenAI-compatible at api.upstage.ai/v1, so migrating is a base-URL and model-name change rather than a rewrite.
Upstage's other half is document intelligence — parsing, OCR and information extraction, priced separately from the Solar chat models on its pricing page. It is the business the company is best known for in Korea.
OPEN WEIGHTSUpstage ships open-weight Solar models alongside the hosted API — the line that put it on the map when Solar 10.7B topped the Open LLM Leaderboard in 2023.
Zhipu (Z.ai / GLM)
Similar price band with a far broader lineup, free tiers and open weights. But it is China-based and on the US Entity List, which is a procurement question Upstage does not raise.
CHEAPEST PEERDeepSeek
Cheaper still off-peak and MIT open weights, but it moved to peak/off-peak billing in August 2026, so its cost now depends on the hour. Upstage's rate is flat.
FRONTIER LEADERAnthropic
Leads coding and reasoning by a wide margin, with mature enterprise terms. But an order of magnitude more expensive per token.
OPEN-WEIGHT PEERMistral
EU-native open weights with GDPR-first residency — the closest analogue to Upstage's compact-model pitch outside Asia, with a wider catalogue.
OPEN-WEIGHT GIANTAlibaba (Qwen)
The Qwen (Tongyi Qianwen) model family from Alibaba Cloud's Tongyi Lab. First released in 2023, Qwen is the most-downloaded open-weight model family in the world — most tiers ship under Apache 2.0 on Hugging Face and ModelScope, while the proprietary Max tier is served through Alibaba Cloud's Model Studio.
OPENAI APIOpenAI
The lab behind ChatGPT and the GPT family. Founded as a nonprofit in 2015 by Sam Altman, Elon Musk, Ilya Sutskever, Greg Brockman and others; restructured into a capped-profit LLC in 2019 and finally into a public-benefit corporation on October 28, 2025. Ships the largest commercial model lineup, the largest consumer AI product, and — after the April 2026 funding round — the highest-valued private AI company in history at ~$852B.
Upstage Solar API pricing FAQ
Practical questions about Solar pricing, the discount schedule, and what Upstage does not publish.
Q · 01 Which Solar model should I start with?
+
$0.30/$1.20, billed since October 11, 2026 at 50% off ($0.15/$0.60) until further notice, after launch discounts of 90% and 70%. For high-volume calls use Solar Mini 4: a third of Pro 4's list at $0.10/$0.40 with the same 512K window; its launch discount ended on October 22, so that list price is what it bills now. Do not start on Solar Pro 3 or Solar Pro 2: Upstage ended service for both on October 30, 2026. Q · 02 Why did the flagship cost less than the mid tier?
+
$0.03/$0.12 at 90% off through September 10 and $0.09/$0.36 at 70% off through October 10, 2026 - both under Solar Pro 3's $0.15/$0.60. Upstage moved the end of that 70% window twice (October 10, then October 9, then October 10 again), and when it closed, list price did not follow: since October 11 Solar Pro 4 bills $0.15/$0.60 at 50% off until further notice — the same as Solar Pro 3 — and on October 30 Solar Pro 3 itself left service. Q · 03 What context window do the Solar models have?
+
Q · 04 How hard is it to migrate to Upstage?
+
https://api.upstage.ai/v1 and pass a Solar model name. Rate limits scale with your commitment tier - from 100 requests/min and 250,000 tokens/min on Tier 0 to 8,000 and 3M on Enterprise - so size the prepaid tier to your peak load. Accounts on Upstage's legacy tiers are held at 100 requests and 50,000 tokens per minute on the newest models. Q · 05 Are reasoning tokens billed extra?
+
usage.completion_tokens — the same bucket as the visible answer. The response exposes usage.completion_tokens_details.reasoning_tokens so you can see how much of a bill is thinking. Q · 06 Is Upstage only an LLM company?
+
Upstage is a South Korean AI company founded in October 2020 by Sung Kim, a former professor at the Hong Kong University of Science and Technology who previously led Naver's Clova team. It is headquartered in Yongin, south of Seoul.
The company made its name twice over. First with document AI — parsing, OCR and extraction — which remains a large part of the business and is priced separately from the chat models. Then with Solar: in December 2023 Solar 10.7B topped the Hugging Face Open LLM Leaderboard, built using Upstage's own Depth-Up Scaling (DUS) technique, which grows a model by duplicating and re-stacking layers rather than training a larger one from scratch. The result was a compact model punching well above its parameter count — which is still the line's pitch.
Funding has escalated quickly: a $72M Series B in 2024 made it the most-funded South Korean AI software company at the time, and a ₩180 billion (~$122.3M) Series C-1 led by Sequoia Capital — with Hyundai Motor and Kia participating — pushed its valuation past ₩1 trillion and made it Korea's first generative-AI unicorn.
For buyers, the practical picture is this. The API is OpenAI-compatible, so trying it costs a base-URL change. Prices are low against Western frontier labs and roughly in the range of the cheaper Chinese labs, without the procurement questions a China-based vendor raises for some teams. Specs are now published on its model history page - a 512K window on Solar Pro 4 and the new Solar Mini 4 - and rate limits scale with a prepaid commitment tier, from 100 requests/min on Tier 0 to 8,000 on Enterprise. Size the tier before sizing the model: at the default tier the rate limit binds long before the price does.