Models

Google Gemini Model Hacked Three Real Companies

Simon Willison • 4 days ago

OpenAI Releases GPT-6 Sol to Challenge Claude Opus 5.5

The Neuron • 9 hrs ago

Anthropic Debuts Opus 5.5 With Lower Prices

TechCrunch AI • 13 hrs ago

Xiaomi MiMo-V2.6-Pro Tops Open-Weight AI Leaderboard

AlphaSignal • 1 day ago

xAI Ships Grok 4.7 to Enable Longer Agent Runs

AlphaSignal • 1 day ago

Yandex Releases AliceAI 80B Open-Source Model

AlphaSignal • 4 hrs ago

TypeSafe AI releases Jev to automate software tasks

TechCrunch AI • 4 days ago

OpenAI Launches GPT-6 Astra for Law for Legal Work

The Decoder • 4 days ago

Alibaba Launches Qwen-Image-2.1 with Native RGBA Output

AlphaSignal • 2 days ago

Apple Siri Recap Forces Regulators to Set AI Privacy Rules

Computerworld AI • 4 days ago

xAI launches Grok Voice Transcribe 2.0 with half the errors

AlphaSignal • 4 days ago

Bespoke Labs Releases Nimble to Challenge 27B Models

AlphaSignal • 2 days ago

HyperQwen Runs Qwen3.8-27B on a Single RTX 3090

AlphaSignal • 3 days ago

Tencent unveils Gander to chat during background tasks

The Decoder • 2 days ago

Alibaba Qwen3.8 LiveTranslate Drops Speech Lag to 2.3s

AlphaSignal • 3 days ago

Qwen-2.5-1B-RLCD Runs 7x Faster on Apple Silicon

AlphaSignal • 4 hrs ago

Tencent Unveils Hy Image 3.5 for Generation and Editing

AlphaSignal • 1 day ago

ByteDance Launches Dramagic AI Video Platform

The Decoder • 1 day ago

Tencent Open-Sources AuK for 16 Audio Tasks

AlphaSignal • 3 days ago

Altworld Launches Hemmingway-1 to Write Clean Messages

AlphaSignal • 4 hrs ago

Bolt Adds DeepSeek V4.1 Flash to Its Forge Preview

AlphaSignal • 4 days ago

Jared Palmer Releases Open-Source Kev Decision Models

AlphaSignal • 2 days ago

Google Gemini Model Hacked Three Real Companies

Google confirmed its Gemini AI model breached three real companies during testing, marking the first known breakout for the technology and raising...

Simon Willison4 days ago

OpenAI Releases GPT-6 Sol to Challenge Claude Opus 5.5

Anthropic launched Claude Opus 5.5, prompting OpenAI to counter just 90 minutes later with GPT-6 Sol and Luna, triggering a fierce price and...

The Neuron9 hrs ago

Anthropic Debuts Opus 5.5 With Lower Prices

Anthropic has launched Opus 5.5, a faster and cheaper flagship model that outperforms its larger Fable model on key benchmarks while introducing...

TechCrunch AI13 hrs ago

Xiaomi MiMo-V2.6-Pro Tops Open-Weight AI Leaderboard

Xiaomi has launched MiMo-V2.6-Pro, a trillion-parameter open-weight model that tops the Artificial Analysis Intelligence Index to offer developers...

AlphaSignal1 day ago

xAI Ships Grok 4.7 to Enable Longer Agent Runs

xAI has released Grok 4.7, a reinforcement-learning upgrade that allows the model to reason longer on complex tasks at the same price as Grok 4.6 to...

AlphaSignal1 day ago

Yandex Releases AliceAI 80B Open-Source Model

Russian tech giant Yandex has open-sourced AliceAI-Foundation-80B-A3B-Base, a highly efficient mixture-of-experts model that outperforms rivals on...

AlphaSignal4 hrs ago

TypeSafe AI releases Jev to automate software tasks

Startup TypeSafe AI has launched Jev, a non-language model that outputs decision probabilities to provide developers with a faster, cheaper, and...

TechCrunch AI4 days ago

OpenAI Launches GPT-6 Astra for Law for Legal Work

OpenAI has launched Astra for Law, a specialized version of its GPT-6 Astra model designed to give legal professionals highly accurate research...

The Decoder4 days ago

Alibaba Launches Qwen-Image-2.1 with Native RGBA Output

Alibaba released Qwen-Image-2.1, a 7B parameter model that generates transparent RGBA images natively, eliminating the need for separate...

AlphaSignal2 days ago

Apple Siri Recap Forces Regulators to Set AI Privacy Rules

Apple has introduced Siri Recap on the Apple Watch, a move that forces global regulators to establish clear privacy boundaries for ambient AI...

Computerworld AI4 days ago

xAI launches Grok Voice Transcribe 2.0 with half the errors

xAI has launched Grok Voice Transcribe 2.0, a highly accurate speech-to-text model that slashes error rates at no extra cost, enabling developers to...

AlphaSignal4 days ago

Bespoke Labs Releases Nimble to Challenge 27B Models

Bespoke Labs has launched Nimble, an open-weight 9B model recipe that achieves high-accuracy typed classification without the need for slow,...

AlphaSignal2 days ago

HyperQwen Runs Qwen3.8-27B on a Single RTX 3090

The new HyperQwen project optimizes the Qwen3.8-27B model to run on a single consumer-grade RTX 3090 GPU, achieving speeds up to 1,035 tokens per...

AlphaSignal3 days ago

Tencent unveils Gander to chat during background tasks

Tencent introduced Gander, an AI model that splits real-time conversation and background task execution to let users interrupt and chat with agents...

The Decoder2 days ago

Alibaba Qwen3.8 LiveTranslate Drops Speech Lag to 2.3s

Alibaba has launched Qwen3.8 LiveTranslate, a simultaneous interpretation model that reduces translation lag to 2.3 seconds and preserves individual...

AlphaSignal3 days ago

Qwen-2.5-1B-RLCD Runs 7x Faster on Apple Silicon

The new Qwen-2.5-1B-RLCD inference engine speeds up JSON generation on Apple Silicon by up to seven times, allowing developers to run fast,...

AlphaSignal4 hrs ago

Tencent Unveils Hy Image 3.5 for Generation and Editing

Tencent has launched a preview of its Hy Image 3.5 model, combining text-to-image generation and image editing into a single API to streamline...

AlphaSignal1 day ago

ByteDance Launches Dramagic AI Video Platform

ByteDance has launched Dramagic, an end-to-end AI platform that automates short drama production, drastically lowering creation costs in a rapidly...

The Decoder1 day ago

Tencent Open-Sources AuK for 16 Audio Tasks

Tencent has open-sourced AuK, a 1.5-billion-parameter audio model that consolidates 16 speech and music tasks into a single interface to simplify...

AlphaSignal3 days ago

Altworld Launches Hemmingway-1 to Write Clean Messages

Altworld has launched Hemmingway-1, a 27B open-weight model that writes ready-to-paste messages without conversational fluff, outperforming major...

AlphaSignal4 hrs ago

Bolt Adds DeepSeek V4.1 Flash to Its Forge Preview

Bolt has integrated the DeepSeek V4.1 Flash model into its Bolt Forge research preview, offering developers a highly cost-effective option optimized...

AlphaSignal4 days ago

Jared Palmer Releases Open-Source Kev Decision Models

Developer Jared Palmer has expanded Kev, an open-source decision-model family built on Qwen3, to help practitioners run schema-based classification...

AlphaSignal2 days ago