Terminal-Bench 2.1
2 stories filed under Terminal-Bench 2.1 on Beyond Market Intelligence. The newest of them: “Nanbeige4.3-33B opens Apache 2.0 access to power smaller, smarter agents” and “Smaller and smarter: Poolside bets transparency beats brute force”. Nanbeige4.3-33B is now open under Apache 2.0, and the numbers tell a focused story. Poolside just shipped Laguna S 2.1, and the numbers deserve a second look. Acme AI is the next-generation, AI-powered spreadsheet platform built to replace Excel and redefine how analysts, data scientists, and enterprise teams work… The list below is every Terminal-Bench 2.1 story on Beyond Market Intelligence, newest first.
Nanbeige4.3-33B opens Apache 2.0 access to power smaller, smarter agents
Nanbeige4.3-33B is now open under Apache 2.0, and the numbers tell a focused story. On Terminal-Bench 2.1, it scores 80.9 against Qwen3-8-27B's 73.0, though it trails 60.3 to 61.7 on SWE-Bench Pro. The team positions this as a teacher model for smaller agents, including a planned 3B model. The 512K context window and preserve_thinking flag for tool use are practical details worth noting. For deeper context on how AI agents are evolving, our coverage of their doubled footprint at NeurIPS offers
Smaller and smarter: Poolside bets transparency beats brute force
Poolside just shipped Laguna S 2.1, and the numbers deserve a second look. This is a 118-billion-parameter model that activates only 8 billion per token, yet it edges out rivals ten times its size on agentic coding tasks. That is not hype; that is a published benchmark score. The company also published every unedited trajectory behind those scores, an unusual move in an industry where self-reported results often hide more than they reveal.