DeepSeek released DeepSeek V4.1 Flash, a 552B-parameter mixture-of-experts model built on a new causal encoder-decoder architecture that natively supports multimodal visual understanding. The model is…
ZenMux, an AI gateway platform, announced a built-in insurance feature that automatically compensates users when AI models produce hallucinated outputs, excessive latency, or low throughput, while als…
OpenAI has released GPT-Image-2.5-Sunburst, a precision-oriented image generation and editing model in the GPT Image 2.5 series, available via the dedicated Images API. The model is priced at $8 per m…
OpenAI released GPT Image 2.5 Flare, an image generation and editing model positioned as the speed-oriented tier of the GPT Image 2.5 series. The model is priced at $8 per million tokens for input and…
OpenAI released GPT-6 Astra, its most capable model for complex reasoning, coding, computer use, research, and document creation, with support for reasoning effort levels from low to max and lower est…
XAI released Grok Imagine Image 2.0, an image generation and editing model that creates images from text prompts and edits images from references, with low and medium quality modes. Pricing is listed …
Meta has released Muse Spark 1.3 Contributor, a cost-efficient tier of its multimodal reasoning model priced at $0.1 per million input tokens and $0.2 per million output tokens, designed for experimen…
Meta released Muse Spark 1.3, a multimodal reasoning model for long-running agentic, multi-agent, and coding workflows, priced at $1.25 per million input tokens and $4.25 per million output tokens. Th…
KlingAI released Kling 3.0 Omni, an all-in-one multimodal model that integrates text, image, audio, and video inputs to enable native audio-visual output. The model is priced at $0.084–0.42 per second…
MiniMax and fal.ai have jointly released MiniMax H3 Max, a video generation model post-trained by fal.ai on MiniMax H3 for high-speed generation, supporting 480P and 768P output. It currently supports…
Anthropic released Claude Fable 5.1, its latest model for demanding reasoning and long-horizon agentic work, featuring a 1M-token context window, adaptive thinking, and support for text and image inpu…
Meta has released Muse Image 1.0, an agentic image generation model that reasons before rendering, breaks down multi-part prompts, and invokes web search for factual accuracy. It supports text-to-imag…
Black Forest Labs released FLUX.2 Pro, a high-end image generation and editing model offering text-to-image and editing up to 4 MP resolution with pricing of $0.015 per megapixel for input and $0.03 f…
Alibaba's Qwen team released Qwen3.8-Flash, a multimodal model with a million-token context window, native support for OpenAI and Anthropic API protocols, and integration with developer tools like Cla…
Z.ai released GLM-5.3-Flash, a native multimodal model designed for efficient coding and long-horizon agent tasks, featuring a hybrid sparse and linear attention architecture that reduces compute over…
Alibaba has released Wan3.0-Video-Prime, a high-speed video generation model that matches the capabilities of the standard Wan3.0-Video, supports four modalities of reference inputs, generates videos …
Alibaba has released Wan3.0-Video, an all-in-one reference-based video generation model supporting text-to-video, image-to-video, and reference-based generation, capable of producing videos up to 30 s…