Flash

TopicKimi K3.1 Pro

Flash·Models & Products·2026-11-17 08:12

Moonshot Pivots Kimi to Science: Biology, Chemistry and Pharma

Moonshot’s international director said at Web Summit in Lisbon on Nov 16 that Kimi models are pivoting from coding to biology, chemistry and pharmaceutical research, as gains on coding benchmarks flatten.

A researcher at a microscope in a sunlit lab
AI-generated illustration

What happened

Giacomo Ficari, Moonshot AI’s international director, said at Web Summit in Lisbon that the company is redirecting its Kimi language models from coding toward biology, chemistry and pharmaceutical research. He said science demands smarter AI models than law, finance or code, and that progress in coding is flattening.

The deployment case: Beijing-based DP Technology is integrating Kimi K3.1 Pro into its research assistant BioMaster. In one test, the system analyzed gene data from 3,696 mouse pancreatic cells and flagged 100 candidate genes in 90 minutes, against an estimated four hours of human work.

The limits: fully independent scientific breakthroughs are still a long way off. On TCSExam, a 100-problem theoretical computer science benchmark, leading models produced verifiable proofs but solved no open conjectures on their own. Anthropic’s Claude found an enzyme system in September, while mathematicians criticized OpenAI’s claimed math and physics breakthroughs; a June Citi survey of 50 US research leaders found 72 percent already using AI broadly, but only 24 percent expected it to noticeably improve drug development success rates.

Key facts

  1. Pivot: Kimi models move from coding to biology, chemistry and pharmaceutical research, per Ficari at Web Summit in Lisbon
  2. Reason: Gains on coding benchmarks are flattening; science demands smarter models than law, finance or code
  3. Case: DP Technology’s BioMaster integrates Kimi K3.1 Pro; gene data from 3,696 mouse pancreatic cells, 100 candidate genes flagged in 90 minutes versus an estimated four hours by hand
  4. Limits: On the 100-problem TCSExam benchmark, models produced verifiable proofs but solved no open conjectures alone
  5. Industry: A June Citi survey of 50 US research leaders: 72% use AI broadly, 24% expect it to noticeably improve drug development success rates

Why it matters

Kimi’s focus moves from coding to scientific research, with DP Technology’s BioMaster as a public deployment; the 90-minutes-versus-four-hours comparison comes from a partner test, and independent scientific discovery remains out of reach.
Useful Tap if this story helped you
Send to a friendMakes a share image for X, WeChat, Telegram and more

SourcesThe Decoder (Nov 16, 2026). This article is compiled from public information and does not constitute investment advice.

From HIMIAO Horizon · himiao.ai

Grab todayOpenAI Codex for Students: $100 in Free CreditsHIMIAO Picks →

Follow HIMIAO Horizon

No sign-up · Unsubscribe any time

Email

Notifications

A ping on your phone or computer when the brief is out. No email needed.

HIMIAO PicksWeekly freebies: one email every Friday, a 5-minute read

Comments

  1. Loading comments…
Ask the cat