Search

Across 58925 posts, found43posts across9collections

SortTitle matches first, then newest

Matching collections

Related editions in the archive

No.004AI Product Updates: New Features and ReleasesUpdated 15 hours ago · Today +014 matches No.006Jev in Practice: Browser Use and Agent WorkflowsUpdated 15 hours ago · Today +010 matches No.020DeepSeek V4.1 Flash hands-on tests and comparisonsUpdated 13 hours ago · Today +05 matches No.005Personal Website Design Inspiration: Portfolios from XUpdated 14 hours ago · Today +05 matches No.012AI Personal Assistants: Muse, Grok Bot, Cue Usage & ComparisonsUpdated 14 hours ago · Today +03 matches No.001GPT-6 Astra in Practice: Tests, Use Cases, and FeedbackUpdated 13 hours ago · Today +03 matches No.008Gemini 4 Pro / Flash: benchmarks, hands-on tests and leaksUpdated 13 hours ago · Today +01 matches No.019Kling 4.0 hands-on tests and comparisonsUpdated 13 hours ago · Today +01 matches No.010Making Videos with Claude: Opus 5.5 Animation and Finished-Film ExamplesUpdated 13 hours ago · Today +01 matches

Matching posts

Matching terms are highlighted

Claude Code experimental agent teams: Opus leads, Haiku swarms

An environment flag enables Claude Code agent teams, where Opus 5.5 plans and signs off, Haiku 5.5 teammates claim parallel subtasks in isolated contexts, and Sonnet 5.5 handles multi-file patches.

claude@0xPontiuz　Filed in No.004

2 days ago

Experimental video clip made with Opus 5.5

The author made an experimental video with Opus 5.5, admits it needs tweaks but is thrilled with the result.

claude@Webilyco　Filed in No.010

4 days ago

WaveSpeed teases Kling 4.0: 3–30s clips, experimental ~120s long-form

Third-party platform WaveSpeed previews Kling 4.0 with 3–30s generations, experimental long-form video up to ~120s, ~10 keyframes, more reference inputs, and Preview plus Flash variants.

kling-4.0@wavespeed_ai　Filed in No.019

2026.09.28

Google launches experimental gaming platform Playground, seen as setup for Argon

Google launched Playground, an experimental US-only 18+ gaming platform where users chat with models to build racing, shooter and platformer games and publish multiplayer titles; the poster suspects it precedes Gemini 4 Argon.

gemini-4@notjazii　Filed in No.008

21 days ago

xAI releases experimental TypeScript SDK for Grok models

One SDK covers Grok text, voice, image and video models and bundles real-time X search, web search, code execution and remote MCP for building multimodal agents.

agent@ai_for_success　Filed in No.004

25 days ago

POST

mcp-handler 2.2.0 adds experimental WebMCP support

Vercel's mcp-handler 2.2.0 lets AI agents browsing your site call WebMCP tools as the signed-in user with no OAuth flow, via an allowlist and a single script tag.

agent@vercel_dev　Filed in No.004

2026.09.19

Jev used to build an experimental Supabase RLS linter

The author built an experimental row-level security policy linter for Supabase with Jev, with a live demo and walkthrough video.

jev@softwarecuddler　Filed in No.006

2026.09.19

POST

Experimental model router for pidotdev uses Jev as its decision maker

Rather than simply picking a cheaper model, it splits work into logical threads and prices cache reads and writes, cold context, expected output and the cost of switching back into its routing decisions.

jev@EngoDev　Filed in No.006

2026.09.19

POST

Experimental context management in Codex cuts Astra quota use

A user reports that enabling experimental context management in Codex noticeably reduces GPT-6 Astra quota consumption, worth trying for those who keep hitting limits.

gpt-6@hqmank　Filed in No.001

2026.09.06

DeepSeek Harness 0.2.1: built-in Search API and V4.1 Flash support

…和 Codex 生态官方新增了 Claude Code 和 Codex 插件组合包，还做了一个实验性的 Claude Code Mods 兼容层不过目前还不是完全兼容，更多是在验证两套插件机制能…

deepseek-v4.1@xiaomovps　Filed in No.020

1 day ago

Showing10of43posts

10 per batch

Keep scrolling to reveal posts 11–20