TLDRocket
Sign in

AI Benchmarking

64 summarised stories about AI Benchmarking, each linking back to the original source. Browse all topics →

+ Follow this topic

Key players

OpenAI Anthropic Hugging Face Google DeepMind Kaggle Sakana AI

Tuesday, 18 August 2026

GPT 5.6 Sol is the best "vision" model OpenAI ever released

Roboflow Blog 1 week ago 16

OpenAI released GPT-5.6 models (Sol, Terra, Luna) with significantly improved vision capabilities, particularly in object detection where Sol achieved 46.2 mAP@50 compared to GPT-5.5's 13.8. Sol costs roughly 2.5 cents per image and takes 10 seconds per image, making it expensive and slow compared to competitors like Gemini 3.5 Flash, though detection and counting are now practical capabilities for document workflows and UI agents.

Read the full story
AI Benchmarking

Cartesia Ships Sonic-3.6: A Streaming TTS Model That Now Leads Both Artificial Analysis Speech Arenas

MarkTechPost 1 week ago 51 ● 2 sources

Cartesia released Sonic-3.6, a real-time text-to-speech model built on state space models rather than transformers. The model ranks first on both Artificial Analysis speech leaderboards with 1,283 Elo on Provider Voice and 1,123 on Controlled Voice, and claims sub-90ms latency to first audio. It is available as a hosted API starting at $5 per month, positioning it at $49 per million characters, half the price of ElevenLabs Eleven v3.

Read the full story
AI Benchmarking Model Release Text-to-Speech

The daily briefing

Every AI story that matters, in your inbox by 8am.

TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.