Qoder IDE Uses AI to Find and Fix 39 Performance Issues, Internal Scores Jump
Qoder IDE reports that AI found and fixed 39 performance issues, with internal scores rising from 45.8 to 84.9 on macOS and 41.4 to 74.5 on Windows between vers…
Global AI Digest
-
-
Command Code Launches VS Code & Cursor Extension in Beta
Command Code released a beta extension for VS Code and Cursor, adding session management, code context, change review, and slash commands to its coding agent li…
-
Qwen Introduces QwenGyre: Elastic RL for Extreme-Long-Horizon Agent Training
QwenGyre brings elastic RL to extreme-long-horizon agent training with an Elastic Scheduler and Trajectory Processor, achieving up to 1.85x speedup and lifting…
-
Tencent TRTC Team Teases Knocket: A Talkable Contact Page Embeddable via a Single Script Tag
Tencent's TRTC team is prototyping Knocket, a free contact page you can talk on, embedded with one async script tag and answered from Telegram, with support for…
-
OpenCode Offers $80 of DeepSeek V4.1 Flash Usage in "Operation Cheepseek Phase 3"
OpenCode announced Phase 3 of its "Operation Cheepseek" campaign, offering $80 of usage credit for DeepSeek V4.1 Flash.
-
Qoder Details 'Fast' Mode Latency and Benchmark Numbers
Qoder published specific latency comparisons for its new Fast mode: time to first response is under 1/3 of Auto mode, under 1/2 of Performance mode, and about 1…
-
Qoder's New 'Fast' Model Cuts Response Time to Under a Third of Auto Mode
Qoder introduced Fast, its new and fastest model, claiming time to first response under 1/3 of Auto mode, under 1/2 of Performance mode, and about 1/4 of Ultima…
-
Command Code GOAT Adds Three GPT Models to Its Lineup
Command Code GOAT now lists three GPT models in its lineup: GPT 6 Luna, GPT 5.6 Sol, and GPT 5.6 Luna.
-
OpenAI Raises GPT-6.1 Sol Default Context to 500k in Codex & Work, Drops Long-Context Fees
OpenAI increased the default context length for GPT-6.1 Sol to 500k tokens in Codex and Work, up from 272k, and removed long-context fees for subscription usage…
-
Command Code Ships Agr, Its First Open Decision Model, Plus Cloud Sandboxes and ACP Support
Command Code's weekly update introduces Agr, described as its first open decision model, alongside cloud sandboxes and ACP support.
