←Search results

Search “GitHub”

By relevance · Total 1623 results Clear search

Thursday

Wednesday

Tuesday

  2. Ben's Bites

    Do you use a personal agent? — Ben's Bites reader survey

    Ben's Bites launches a reader survey asking who is using "personal agent" products—that is, tools that help organize affairs, handle email, and perform actions on your computer outside of work; the answers will be revealed in Thursday's article. This issue also mentions that Codex/ChatGPT has added a "computer history" feature that can turn desktop app and website activity into memories and a timeline, which must be enabled in settings; Claude Code has added a /design skill that can generate editable UI canvases and implement them directly.

    Ben's Bites launches a reader survey asking who is using "personal agent" products—that is, tools that help organize affairs, handle email, and perform actions on your computer outside of work; the answers will be revealed in Thursday's article. This issue also mentions that Codex/ChatGPT has added a "computer history" feature that can turn desktop app and website activity into memories and a timeline, which must be enabled in settings; Claude Code has added a /design skill that can generate edi

    Matched fieldsBody

    #newsletter#Industry news#Agents#OpenAI

Saturday

Friday

  1. Ant Ling: Developer Blog

    Ant Ling closes the single-machine Agentic RL loop with Ling-3.0-tiny and AReno

    Ant Ling has published a practice article describing how it used the local post-training toolkit AReno on a DGX Spark to perform Agentic RL on Ling-3.0-tiny, using tic-tac-toe as a minimal validation task to complete the four-step loop of baseline evaluation, Reward definition, post-training, and re-evaluation in the same environment. The experiment used GSPO for 400 training steps; rollout/rewards_mean rose from about -0.5 to around 0.4, response_len dropped to about 850 tokens, and tool calling and move selection became more stable in the re-evaluation. Ling-3.0-tiny is a hybrid reasoning model with 7.9B total parameters and 1.3B activated parameters; BF16, FP8, and INT4 versions have been open-sourced, and AReno provides a Docker reproduction tutorial.

    English text is being prepared.

    Ant Ling has published a practice article describing how it used the local post-training toolkit AReno on a DGX Spark to perform Agentic RL on Ling-3.0-tiny, using tic-tac-toe as a minimal validation task to complete the four-step loop of baseline evaluation, Reward definition, post-training, and re-evaluation in the same environment. The experiment used GSPO for 400 training steps; rollout/rewards mean rose from about -0.5 to around 0.4, response len dropped to about 850 tokens, and tool callin

    Matched fieldsBody

    #engineering#Tutorials / practice#Agents#Open-source ecosystem

Thursday

Wednesday

Tuesday

Monday