Fresh stories
Codex opens 1M-token GPT-5.6 Sol context for ChatGPT subscribers
Codex now lets ChatGPT subscribers enable a 1 million-token context window for GPT-5.6 Sol. Automatic compaction starts at 900,000 tokens, and tokens beyond the default window count double against limits.
Reports: GLM-5.3 post-training lifts Terminal-Bench from 4.6 to 28.3
Reports say GLM-5.3 retained GLM-5.2's base model while post-training raised Terminal-Bench from 4.6 to 28.3 and DeepSWE from 46.2 to 66.9. A technical account attributes the gains to RL infrastructure changes.
OpenRouter updates Auto Router with 30 task types and 7-day spend-based routing
OpenRouter upgraded Auto Router to classify prompts into about 30 task types, then route by anonymized 7-day spend share and cost tier. OpenRouter says the max tier beat the old router across five benchmark domains.
OpenCode Go adds $30 usage credit to its $10 plan
OpenCode revised its Go plan to include $30 in usage for a $10 subscription. The change follows a provider price increase and the company’s effort to secure lower-cost capacity.
Codex opens 1M-token GPT-5.6 Sol context for ChatGPT subscribers
Codex now lets ChatGPT subscribers enable a 1 million-token context window for GPT-5.6 Sol. Automatic compaction starts at 900,000 tokens, and tokens beyond the default window count double against limits.
OpenCode Go revises limits after DeepSeek price increase
OpenCode says it revised Go limits after DeepSeek raised prices. Its operator is testing hosting configurations intended to bring DeepSeek service closer to its prior price point.
Reports: GLM-5.3 post-training lifts Terminal-Bench from 4.6 to 28.3
Reports say GLM-5.3 retained GLM-5.2's base model while post-training raised Terminal-Bench from 4.6 to 28.3 and DeepSWE from 46.2 to 66.9. A technical account attributes the gains to RL infrastructure changes.
Qwen 122B runs on older laptop with llama.cpp in user test
Composio and Ante benchmark coding agent harnesses with 47%–67% success range
OpenRouter updates Auto Router with 30 task types and 7-day spend-based routing
Microsoft Copilot traces report 87% of LLM calls came from agents
Briefs forAugust 17
Daily AI Digest
Get the best stories delivered
to your inbox