Timeline
Claude Code 2.1.224 removes the old 200-subagent cap, enabling truly parallel workflows.
Scored 78.9% on Terminal-Bench 2.1 and 88.6% on SWE-bench Verified with Opus 4.6
Released version 2.1.221 with credential masking, VSCode Focus view, and 20 security fixes
Claude Code shifts to a policy-controlled execution layer with deterministic security workflows
Shift to a policy-controlled execution layer in the harness, enforcing deterministic security workflows
Open-source plugin 'dont-let-me' released for Claude Code, Codex, and OpenCode
Ollama integrates support for Codex with DeepSeek V4, Gemma 4, Qwen 3.6 for local execution
Benchmark revealed it collapsed under load of 5 concurrent users, highlighting gap between developer-friendly tools and production-ready systems.
Ollama expands its service to include cloud-hosted model deployment, starting with MiniMax's M2.7.
Added support for Apple's MLX framework as a backend for local LLM inference on macOS
Ecosystem
Claude Code
Llama
Benchmarks
Evidence (6 articles)
How to Run Claude Code Locally with Ollama for Free, Private Development
Mar 25, 2026How to Keep Coding When Claude Code Goes Down: Your Local Fallback Plan
Mar 18, 2026WSL 3 Preview: Cut Claude Code's Local Inference Latency on Windows
Jun 23, 2026How to Run Claude Code on Local LLMs with VibePod's New Backend Support
Mar 18, 2026Amazon's SageMaker Agentic Fine-Tuning Supports Llama, Qwen, DeepSeek, Nova
May 5, 2026Build a Claude Code Fallback Chain
Jul 29, 2026