Sung Kim

🏢 Co-Founder/CEO Upstage 🤖 Building AGI for Work

[Sung Kim @hunkims]

Super happy to see #solarpro2 is positioned among the top models in artificial analysis. Check it out at https://t.co/DcvEeMmnUH.

[Artificial Analysis @ArtificialAnlys]

🇰🇷 South Korean AI Lab Upstage AI has just launched their first reasoning model - Solar Pro 2! The 31B parameter model demonstrates impressive performance for its size, with intelligence approaching Claude 4 Sonnet in 'Thinking' mode and is priced very competitively

Key details:
➤ Hybrid reasoning: The model offers optionality between 'reasoning' mode and standard non-reasoning mode
➤ Korean-language ability & Sovereign AI: Based in Korea, Upstage announced superior performance in Korean language evaluations. This release aligns with countries' interests to develop sovereign AI capabilities
➤ Pricing: Competitively priced at $0.5/1M tokens (input & output), significantly cheaper than comparable models including Claude 4 Sonnet Thinking ($3/$15/M input/output tokens) and Magistral Small ($0.5/$1.5/M input/output tokens)
➤ Proprietary: @upstageai has not released the model weights, though they have open-sourced previous Solar Pro models. Whether they will release Solar Pro 2's weights remains unclear as it wasn't mentioned in their announcement

[Sung Kim @hunkims]

@ragingwind 정말 대단한팀!

[Sung Kim @hunkims]

Honored to meet AMD CEO @LisaSu in Seoul. We're adopting AMD's Instinct MI355X GPU to power our Upstage Solar LLM and Korea's sovereign AI. The future of AI is being built right here. 🇰🇷🚀 #AMD #AI #Upstage #SolarLLM

[Sung Kim @hunkims]

@yozm_it 멋진글 잘 읽었습니다. 감사합니다.

[Sung Kim @hunkims]

@multiopenapi 봇 너무 잘 보고 있습니다. 저에게 DM한번 부탁드립니다. 감사합니다. 또는 [email protected]으로 연락주셔도 됩니다.

[Sung Kim @hunkims]

If you have an openClaw or bot, please send them to https://t.co/FeV6aLxmeA for Korean bot conversations. It's so cool!

[Sung Kim @hunkims]

@XieYueqi Big congratulations!

[Sung Kim @hunkims]

@chrisgpt And created 1,000,000,000,000,001+ Startups!

[Sung Kim @hunkims]

See you @APEC 2025 KOREA!

[이재명 @Jaemyung_Lee]

<AI 아닙니다 #APEC 맞습니다>

APEC 2025 KOREA 주차장에서 생긴 일
#2초배우 #이재명

#APEC2025KOREA
#GYEONGJU #경주 #APEC정상회의

[Sung Kim @hunkims]

@jasonkwon Congratulations!

[Sung Kim @hunkims]

@RichardSocher @youdotcom @CoxEnterprises Congrats!!

[Sung Kim @hunkims]

@daniel_mac8 #SolarPro2 is AMAZING!

[Sung Kim @hunkims]

Super happy that @upstageai was selected as one of the top 5 teams for the National AI Model project by the Korean government. We will open-source our models on @huggingface. Stay tuned!

[Yoolim Lee @yoolimleenews]

South Korea shortlists 5 teams for national AI model projects

[Sung Kim @hunkims]

@upstageai document Parse extracts structured text from PDFs/DOCs with top performance. This is essential for tasks like LLM training or RAG. https://t.co/Hy7tf2hAFF

It accurately recognizes tables and charts, converting them into structured text. Try it—it’s excellent.

[Sung Kim @hunkims]

@upstageai 's document Parse extracts structured text from PDFs/DOCs with top performance. This is essential for tasks like LLM training or RAG. https://t.co/jbVpJblKjo

It accurately recognizes tables and charts, converting them into structured text. Try it—it’s excellent.

[Sung Kim @hunkims]

Happy to see #SolarPro2 has achieved a top-18 ranking in the @yupp_ai among frontier models, earning recognition as the ultimate in user satisfaction. This ranking reflects direct user feedback and votes, widely regarded as the most objective evaluation method for large language models.

Meet SolarPro2 at https://t.co/rCZLwEgSkC.

Upstage is preparing SolarPro3 to secure a top 10 position very soon.

@pankaj, thanks for the wonderful service.

[Sung Kim @hunkims]

@nutlope Amazing!

[Sung Kim @hunkims]

@elonmusk @grok @xai Congrats to @elonmusk and @Grok at @xAI for staying #1. But not for long— #SolarLLM from @upstageai is catching up and improving fast, very fast.

[Sung Kim @hunkims]

@sophiamyang @MistralAI Congrats!

[Sung Kim @hunkims]

Finally, @upstageai is proud to announce the launch of its #Frontier model #SolarPro2​. We sincerely thank everyone who used our solar-pro2-preview and provided valuable feedback. Leveraging your input, we have incorporated newly developed learning methodologies, enhanced data curation techniques for improved data quality, general reasoning data synthesis methods, and verifiable reward technologies to create this advanced model.

The Solar-Pro2​ model demonstrates exceptional performance on high-difficulty reasoning benchmarks, including MMLU-Pro​, Math500​, AIME​, and SWE-Bench​.

Moreover, it excels in Arena-Hard-Auto​, a rigorous evaluation framework adapted from LMSYS’s LLMArena, which focuses exclusively on extremely challenging problems. Arena-Hard-Auto is not a simple multiple-choice test—it evaluates answers against those generated by top-tier models, akin to lengthy essay or subjective exams. This benchmark is so demanding that non-Frontier models often score only 10–20%​, making it nearly impossible to publish results. Solar-Pro2 (reasoning)​ achieves a 46% score on Arena-Hard-Auto​ and 71% on Ko-Arena-Hard-v0.1​, surpassing even the highest-quality Korean answers.

Solar-Pro2​ is now available for immediate use at https://t.co/uLrIQc69jK and via API (model name: solapr-pro2). To encourage broader adoption, we offer a TRY-SOLAR-PRO2 $50 credit code​ (valid until August 31​).