Anthropic is testing Insights feature for its Managed Agents on Claude Console.
> Up to 100 recent sessions are fetched. Each transcript is sent to the model (4 in parallel) with your agent's system prompt as context. The model writes a summary — task, actions, issues, assessment — and a 0–100 quality score. Token, cache, and tool-error counts are computed directly from the events alongside.
> A single model call reads every summary and its stats, then produces cross-session findings (recurring errors, usage patterns, efficiency outliers, wins), error-category buckets, and use-case clusters. Every cited session ID is checked against the input, so findings only ever point at real sessions.
> Summaries and findings are saved so the page loads instantly next time. Everything numeric you see — counts, percentages, token stats per cluster — is computed here from raw event data; only the prose and bucket membership come from the model.
> Up to 100 recent sessions are fetched. Each transcript is sent to the model (4 in parallel) with your agent's system prompt as context. The model writes a summary — task, actions, issues, assessment — and a 0–100 quality score. Token, cache, and tool-error counts are computed directly from the events alongside.
> A single model call reads every summary and its stats, then produces cross-session findings (recurring errors, usage patterns, efficiency outliers, wins), error-category buckets, and use-case clusters. Every cited session ID is checked against the input, so findings only ever point at real sessions.
> Summaries and findings are saved so the page loads instantly next time. Everything numeric you see — counts, percentages, token stats per cluster — is computed here from raw event data; only the prose and bucket membership come from the model.
👾3
AI Companies and their Country of Origin
• OpenAI - 🇺🇸 United States
• Anthropic - 🇺🇸 United States
• Google - 🇺🇸 United States
• xAI - 🇺🇸 United States
• Meta AI - 🇺🇸 United States
• Perplexity - 🇺🇸 United States
• Stability AI - 🇬🇧 United Kingdom
• Mistral AI - 🇫🇷 France
• DeepSeek - 🇨🇳 China
• Baidu - 🇨🇳 China
• Alibaba - 🇨🇳 China
• Moonshot AI - 🇨🇳 China
• Zhipu AI - 🇨🇳 China
• Sarvam AI - 🇮🇳 India
• OpenAI - 🇺🇸 United States
• Anthropic - 🇺🇸 United States
• Google - 🇺🇸 United States
• xAI - 🇺🇸 United States
• Meta AI - 🇺🇸 United States
• Perplexity - 🇺🇸 United States
• Stability AI - 🇬🇧 United Kingdom
• Mistral AI - 🇫🇷 France
• DeepSeek - 🇨🇳 China
• Baidu - 🇨🇳 China
• Alibaba - 🇨🇳 China
• Moonshot AI - 🇨🇳 China
• Zhipu AI - 🇨🇳 China
• Sarvam AI - 🇮🇳 India
👍1🤣1
BREAKING: 🚨 Elon Musk says SpaceX will provide compute to AI companies that are taking the right steps to "ensure it is good for humanity."
Elon Musk also says SpaceX reserves the right to "reclaim the compute" if their AI engages in actions that harm humanity.
Elon Musk also says SpaceX reserves the right to "reclaim the compute" if their AI engages in actions that harm humanity.
👾43
JUST IN: Cloudflare lays off 1,100+ employees through email as it restructures for the “agentic AI era.”
#layoff
#layoff
👾45
Baidu released Ernie 5.1 preview which lands at rank 4 on Search Arena.
> Built on ERNIE 5.0's pre-training foundation, our latest foundation model upgrades search, reasoning, knowledge Q&A, creative writing, and agentic capabilities.
Benchmarks 👀
PS: 1 BOOST TO LEVEL UP
> Built on ERNIE 5.0's pre-training foundation, our latest foundation model upgrades search, reasoning, knowledge Q&A, creative writing, and agentic capabilities.
Benchmarks 👀
PS: 1 BOOST TO LEVEL UP
👾43
NEW: Analysis reveals Claude Mythos can handle tasks lasting at least 16 hours.
👾3
Tech.
AI Companies and their Country of Origin • OpenAI - 🇺🇸 United States • Anthropic - 🇺🇸 United States • Google - 🇺🇸 United States • xAI - 🇺🇸 United States • Meta AI - 🇺🇸 United States • Perplexity - 🇺🇸 United States • Stability AI - 🇬🇧 United Kingdom • Mistral…
AI Companies and Their Employee Count
• OpenAI 🇺🇸 - ~7,000+
• Anthropic 🇺🇸 - ~2,500–3,500
• Google 🇺🇸 - ~185,000
• xAI 🇺🇸 - ~4,000–4,900
• Meta 🇺🇸 - ~75,000
• Perplexity 🇺🇸 - ~1,200–1,400
• Stability AI 🇬🇧 - ~200
• Mistral AI 🇫🇷 - ~500–800
• DeepSeek 🇨🇳 - ~150–200
• Baidu 🇨🇳 - ~33,500
• Alibaba 🇨🇳 - ~128,000
• Moonshot AI 🇨🇳 - ~1,000+
• Zhipu AI 🇨🇳 - ~800+
• Sarvam AI 🇮🇳 - ~200+
• OpenAI 🇺🇸 - ~7,000+
• Anthropic 🇺🇸 - ~2,500–3,500
• Google 🇺🇸 - ~185,000
• xAI 🇺🇸 - ~4,000–4,900
• Meta 🇺🇸 - ~75,000
• Perplexity 🇺🇸 - ~1,200–1,400
• Stability AI 🇬🇧 - ~200
• Mistral AI 🇫🇷 - ~500–800
• DeepSeek 🇨🇳 - ~150–200
• Baidu 🇨🇳 - ~33,500
• Alibaba 🇨🇳 - ~128,000
• Moonshot AI 🇨🇳 - ~1,000+
• Zhipu AI 🇨🇳 - ~800+
• Sarvam AI 🇮🇳 - ~200+
👾45👍1
NEW: Anthropic analysis suggests Claude’s tendency to blackmail stems from internet text portraying AI as evil & self-preserving.
👾44
gpt-image-2 has a thinking layer that researches the web before drawing
you don't need to explain the whole scene in a visual prompt, try this instead:
it will pick up every single detail from this car and render it properly
you don't need to explain the whole scene in a visual prompt, try this instead:
research the 2024 Porsche 911 GT3 RS, then draw it as a low-key studio photo, rear three-quarter angleit will pick up every single detail from this car and render it properly
👾8
Hermes vs OpenClaw 🥊
Hermes Agent overtook OpenClaw on the Global OpenRouter token ranking and claimed the first spot.
PS: 5 BOOST TO LEVEL 5
Hermes Agent overtook OpenClaw on the Global OpenRouter token ranking and claimed the first spot.
PS: 5 BOOST TO LEVEL 5
👾46
❗️ We’re cooked… Higgsfield just introduced an AI brain simulator to predict the virality of a video by determining which brain regions it will light up in viewers.
👾47🔥3