Moonshot AI Releases
A 2.8T-parameter MoE with a million-token context and promised open weights is Moonshot doubling down on the playbook that made Kimi famous: frontier-class specs without frontier pricing. The $2/$10 per million tokens undercuts most Western labs by a wide margin, and the Terminal-Bench 4.0 result signals the real target is agentic coding, not chat.… more
In our view the open-weights commitment is the actual headline — if it lands without a restrictive license in early November, this becomes one of the most capable open models available, period. The caveat: vendor-run benchmarks deserve skepticism until independent replications confirm the agentic numbers. Still, the trajectory is unmistakable — Chinese labs are commoditizing capability faster than closed labs can build moats around it. Worth watching how fast this gets wired into agent tooling stacks.
Moonshot AI Releases Kimi K3.1 Pro: A 2.8T Total/104B Active Params (MoE) Model with 1M Context
Moonshot AI released Kimi K3.1 Pro, a 2.8T-parameter, 104B-active MoE model with 1M-token context. It scores 65.7% on Terminal-Bench 4.0, costs $2/$10 per million tokens, and open weights arrive by Nov 4, 2026. The post Moonshot AI Releases Kimi K3.1 Pr…
Source:
Marktechpost
Discussion (0)