BREAKING: Mistral AI has released new OCR models – mistral-ocr-2503 and mistral-ocr-latest. Mistral's OCR features are free to try on Le Chat. To test the API, visit La Plateforme.
SOFTWARE
-
Claude 3.7 Sonnet released with extended thinking budget control
By
–
We've also added support for our latest model, Claude 3.7 Sonnet, and the ability to control the extended thinking budget—right from the Workbench: https://
anthropic.com/news/upgraded-
anthropic-console
… -
AI Service Meshes: A New Frontier In AI Observability
By
–
AI Service Meshes: A New Frontier In AI Observability
#AI #AIio #BigData #ML #NLU #Futureofwork @HaroldSinnott @fogoros @iainljbrown @NandoDF @katecrawford @drhassanrashidi @YuHelenYu http://
ow.ly/Fm4G30sK3lH -
Database Optimization Speeds Up Agent-Powered Applications
By
–
Our databases just keep getting better. They're now faster than ever, optimized to reduce latency—especially for apps built with Agent. You'll notice more snappiness and better apps thanks to our optimizations (and our amazing partners at Neon, of course)
-
Flowise Now Supports MCP Protocol for LLM Integration
By
–
Today, we're supporting MCP in Flowise!
— FlowiseAI (@FlowiseAI) 6 mars 2025
There has been a lot of discussion about MCP lately. From our perspective, MCP is nothing fancy – just tools with third-party APIs or functions, wrapped under a unified protocol that can be consumed by client LLM applications.
In fact,… pic.twitter.com/XXiOadDRszToday, we're supporting MCP in Flowise! There has been a lot of discussion about MCP lately. From our perspective, MCP is nothing fancy – just tools with third-party APIs or functions, wrapped under a unified protocol that can be consumed by client LLM applications. In fact,
-
Flowise AI Opening Public Access Next Month
By
–
send us an email to support@flowiseai.com, or actually we are opening public access next month!
-
Integration platforms: owned clients versus community-driven approaches
By
–
Do you think there will be a lot of clients that use them? Or one that goes deep on creating a lot of integrations, and then also owns the client Eg Zapier owns the client, I think they maintain core integrations, and then there’s community ones
-
Codeium Windsurf Wave 4 launches with AI preview and automation features
By
–
Codeium released Windsurf Wave 4 with new features, including:
— Rowan Cheung (@rowancheung) 6 mars 2025
—AI previews for rapid app iteration
—Tab-to-import functionality
—Suggested actions
—Linter integration
—Model choices for enterprise adminspic.twitter.com/DWDDnA66N7Codeium released Windsurf Wave 4 with new features, including: —AI previews for rapid app iteration
—Tab-to-import functionality —Suggested actions
—Linter integration
—Model choices for enterprise admins -
Alibaba QwQ-32B: Reasoning AI Matches DeepSeek-R1 Cheaper
By
–
Alibaba’s Qwen team dropped QwQ-32B, a reasoning AI that matches or surpasses DeepSeek-R1 at much less cost
— Rowan Cheung (@rowancheung) 6 mars 2025
—20x smaller than DeepSeek-R1
—Priced $0.20 per million input and output tokens
—Open-sourced under Apache 2.0
Here it is running on M4 Max:pic.twitter.com/J40JG0jGtCAlibaba’s Qwen team dropped QwQ-32B, a reasoning AI that matches or surpasses DeepSeek-R1 at much less cost —20x smaller than DeepSeek-R1
—Priced $0.20 per million input and output tokens
—Open-sourced under Apache 2.0 Here it is running on M4 Max: -
HeadInfer: Long-Context LLM Inference on Consumer GPUs
By
–
HeadInfer: Unlocking Long-Context LLM Inference on Consumer GPUs (Million-level Tokens)
*long-context inputs require large GPU memory.
*A standard LLM like Llama-3–8B requires 207GB of GPU memory for 1 million tokens — far beyond the capabilities of consumer GPUs like the RTX