Kimi released K2.6, an open-source model advancing autonomous coding and long-horizon execution. The model demonstrates improvements in coding tasks across multiple languages and is available via API, IDE, and apps. Demonstrated capabilities include 12+ hour autonomous executions with 4,000+ tool calls, including optimizing model inference and refactoring complex codebases.
Models
Kimi K2.6: Advancing Open-Source Coding
Kimi open-sources K2.6, demonstrating autonomous coding with 12+ hour executions and 4,000+ tool calls, handling complex refactoring and optimization tasks across multiple languages.
Monday, April 20, 2026 12:00 PM UTC2 MIN READSOURCE: Hacker NewsBY sys://pipeline
Tags
models
/// RELATED
StrategyApr 27
Microsoft and OpenAI gut their exclusive deal, freeing OpenAI to sell on AWS and Google Cloud
OpenAI escapes Microsoft's exclusivity trap, gaining freedom to sell AI services on AWS and Google Cloud—a seismic shift in cloud AI distribution.
Models5d ago
Granite 4.1 LLMs: How They’re Built
IBM's Granite 4.1 open-source LLM family (3B–30B params, 512K context) achieves superior performance to its larger 32B MoE predecessor through dense architecture and multi-stage training on 4.1M curated samples with GRPO reinforcement learning.