@magnitudedev/cli@0.2.1
vadimdemedes/ink@magnitudedev/cli@0.2.1Sep 30, 2026by github-actions[bot]
Upgrading?
AI Summary
A patch release focusing on concurrency issues, specifically fixing local model hanging when new requests arrive during generation and resolving speculative decoding crashes on long prompts.
Key Highlights
- Fixed local model hanging forever when a new request arrives while another is generating.
- Fixed speculative-decoding models failing mid-request with 'blocked last page' errors.
- Improved engine admission logic to prevent waiting on running generations indefinitely.
New Features
- Enhanced concurrency handling for the inference engine.
- Improved stability for speculative decoding on long context lengths.
Full Release Notes
## 0.2.1 ### Patch Changes - [`0476b71`](https://github.com/magnitudedev/magnitude/commit/0476b71a0fb772245328529312a3392d77b630aa) Thanks [@anerli](https://github.com/anerli)! - - Fix the local model hanging forever when a request arrived while another was generating, as with Qwen3.6 35B-A3B: every later request waited without a response while the model still reported Ready, and the engine held a CPU core at 100%. Admission no longer waits on the running generation, and the engine can no longer wait on work only it could release. - Fix speculative-decoding models failing mid-request with "only a blocked last page is relocated" during long prompts.
Related Repositories
openclaw/openclaw★ 392,714
The AI that really does things. Any OS. Any Platform. The lobster way. 🦞
nilbuild/developer-roadmap★ 370,095
Interactive roadmaps, guides and other educational content to help developers grow in their careers.
morluto/rea★ 360,153
Reverse engineer anything with agents, from app behavior down to native binaries.
anomalyco/opencode★ 219,855
The open source coding agent.
n8n-io/n8n★ 208,278
Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.