Grinding on @openclaw , updated to 2026.3.8, next line item is to set up fallbacks; models that will take over in the even the local @ollama is too slow or times out; @OpenRouter is going to be secondary (with about $10 in credits), and of course @xai @grok tertiary.
#openclaw #openrouter #xai
Я дал ИИ собственный компьютер и 483 сессии свободы. Вот что произошло Эксперимент в автономии искусственног...
#AI #llm #openrouter #qwen #ИИ #сознание #философия #эксперимент
Origin | Interest | Match
Jak używamy modeli językowych? 🤔 Raport OpenRouter pokazuje m.in. Glass Slipper Effect, agentic inference i role-playing. Więcej wniosków z tego badania znajdziesz na naszym blogu: azurro.pl/raport-openr...
#AI #AgenticAI #OpenRouter #AIInsights
....i dzisiaj na drugą nóżkę kontynuacja wczoraj linkowanego artykułu, tym razem o podpięciu równie modnego #OpenRouter pod #OpenClaw. Ponownie Tomek Dunia w akcji i ponownie tutorialowa forma.
#AI #SztucznaInteligencja
blog.tomaszdunia.pl/openrouter/
I’ve shipped: #OpenRouter Provider for #WordPress
Connect your WP site to 200+ AI models (GPT-5.2, #Claude, Gemini & more) through a single API key.
Built on the official PHP AI Client SDK for #WordPress 7.0+
Open source & free:
github.com/zaherg/ai-pr...
#WordPress #AI #OpenRouter #PHP
OpenRouter Review: One API Key to Rule Them All
https://awesomeagents.ai/reviews/review-openrouter/
#Openrouter #Api #Inference
Playing with #nanobot I started using #openrouter (as recommended), but openrouter is turning out to be ridiculously unreliable.
Свой ChatGPT бот в Telegram в 2026 с генерацией изображений, зрением и веб-поиском Пришло время обновить бота. В этой с...
#chatgpt #openai #telegram #bot #proxyapi #openrouter
Origin | Interest | Match
Qwen3.5: Towards Native Multimodal Agents Qwen3.5: Towards Native Multimodal Agents Alibaba's Qwen just released the first two models in the Qwen 3.5 series - one open weights, one proprietary....
#ai #generative-ai #llms #vision-llms #qwen […]
[Original post on simonwillison.net]
Por cierto, si alguno no se lo ha planteado... Aunque se viene #Sonnet5, ya es buena hora para pasarse a modelos locales (siempre que tengan el hardware necesario) o con opciones tipo #openrouter.
Y no, mas consumo de tokens o mas codigo generado no equivale a mejor código.
I had the chance to play with OpenClaw (Clawdbot) today. I used OpenRouter and one of the free LLMs for the model, with chat over Slack. When I couldn't config Slack for the bot, it fixed the config itself and restarted its own service. Interesting times!
#clawdbot #openclaw #openrouter #AI
Al in one ai aizolo.com
Al in one ai aizolo.com
#ai #allinoneai #allinone #aizolo #openai #gemini #claude #kimi #deepseek #openrouter #imagegenerator #aivideogenerator
just released a new version of my #crypto #trading #ai bot for #kraken exchange. the premise is that it feeds your portfolio data to the AI (secure, through #openrouter) which then gives "advices" which is (MCP style) executed by the bot. https://github.com/yonie/kraken-bot
RE: https://mstdn.feddit.social/@admin/115909921287682040
再换成了deepseek/deepseek-chat，直接给我打了十几个tag，何意为......
然后AI Tag再次失败的原因是没有打开“Enable paid endpoints that may train on inputs”
#deepseek #karakeep #openrouter #pricacy
RE: https://mstdn.feddit.social/@admin/115909673373377292
免费的才是最贵的，换成付费模型，虽然不贵，但是效果还是不错的，速度比免费模型快几倍
INFERENCE_TEXT_MODEL=openai/gpt-oss-120b
INFERENCE_IMAGE_MODEL=openai/gpt-4.1-nano
#openrouter #openai #karakeep #gpt_oss_120b #gpt_4.1_nano
privacy
.env
Selfhosted project categories
kiwix
用openrouter的免费模型给karakeep打tag，需要在 https://openrouter.ai/settings/privacy 允许两个权限
不过library.kiwix.org开了anibus好像就无法打AI Tag
#karakeep #bookmark #openrouter #tag
OpenRouter now lets you run Claude Code with 39 free models via 3-line config—will you switch or stick to official?
#OpenRouter #ClaudeCode
open.substack.com/pub/aidisrup...
Сделай бота для работы Согласно Hype Cycle от Gartner за 2025 год, AI-агенты достигли пика завышенных ожиданий. Но что с...
#ai-agent #ai-ассистент #ai-агент #ai-tutorial #нейросети #llm #openrouter #telegram #транскрибация #сезон #ии
Origin | Interest | Match
Using #GitHub actions for scheduled jobs is pretty cool. I now have 2 repos running scheduled jobs, fetching and categorizing news using LLMs, sending daily summaries to a #telegram group. Combined with the #openrouter free tier this is a great way to build simple auto-updating data pipelines
🔧 #OpenRouter launches Response Healing: automatically fixes malformed #JSON responses from #LLMs before they reach your app #AI #API #devops
📊 Results from 1.6M+ requests analyzed at inference time:
• #Gemini 2.0 Flash: 80% defect reduction (99.61% → 99.92%)
🧵 👇
🔍 Schema-aware query intelligence through secure introspection understands your database structure for accurate query generation
🔌 Multi-model flexibility: supports #OpenAI, #Anthropic, #Gemini, #OpenRouter compatible APIs, and fully local inference via #Ollama
Want to use #AI models from OpenAI, Anthropic, Mistral, etc. in #ONLYOFFICE with a single API key?
Our new #OpenRouter integration makes it possible. Enhance your workflow with AI features in DocSpace and Desktop Editors.
🔗 onlyo.co/4oMKRR4
The release of the first widely adopted reasoning model, #o1, marked a #turningpoint in the evolution of #LLMs. An empirical #study using the #OpenRouter platform analysed over 100 trillion tokens of real-world LLM interactions, revealing substantial adoption of #openweightmodels, the popularity of…
Enjoying @EclipseFdn #TheiaIDE for desktop + #Cline + #OpenRouter … I have a good vibe about this 😉
The following features are included:
- auth to bsky using app key
- retrieve posts from a bsky list
- perform filtering of posts (ie date, topics)
- network graph of conversation topics/users
- choose a model on #OpenRouter to use
- use llm to analyze sentiment/views and persuasive techniques
I’ve been having fun lately with #openrouter and #cline. Any good model you would recommend (beyond the “classic” Claude Sonnet)?
I’ve been having fun lately with #openrouter and #cline. Any good model you would recommend (beyond the “classic” Claude Sonnet)?
Kimi K2 Thinking Kimi K2 Thinking Chinese AI lab Moonshot's Kimi K2 established itself as one of the largest open weight models - 1 trillion parameters - back in July. They've now released ...
#ai #generative-ai #llms #llm #pelican-riding-a-bicycle […]
[Original post on simonwillison.net]
Kimi K2 Thinking Kimi K2 Thinking Chinese AI lab Moonshot's Kimi K2 established itself as one of the largest open weight models - 1 trillion parameters - back in July . They've now released...
#ai #generative-ai #llms #llm #pelican-riding-a-bicycle #llm-reasoning #llm-release #openrouter […]
