Baseten
PulseAugur coverage of Baseten — every cluster mentioning Baseten across labs, papers, and developer communities, ranked by signal.
- 2026-06-26 funding AI inference startup Baseten secured $1.5 billion in Series F funding, reaching a $13 billion valuation. source
- 2026-06-20 funding Baseten is reportedly close to closing a $1.5 billion funding round at a $13 billion valuation. source
- 2026-06-18 funding Baseten is reportedly raising $1.5 billion at a $13 billion valuation. source
- 2026-06-18 funding AI inference startup Baseten is reportedly raising $1.5 billion at a $13 billion valuation. source
- 2026-06-18 funding AI inference startup Baseten is reportedly nearing a $1.5 billion funding round at a $13 billion valuation. source
11 day(s) with sentiment data
-
Fireworks AI highlights ecosystem collaboration for legal AI model training
Fireworks AI is highlighting its role in the AI ecosystem by reposting a message from Harvey. The message details how various entities, including Fireworks AI, are collaborating to post-train models for legal work. This…
-
NVIDIA launches Nemotron 3.5 Lightning for efficient agentic AI · 10 sources tracked
NVIDIA has launched Nemotron 3.5 Lightning, a 30-billion-parameter mixture-of-experts model designed for efficient, high-volume agentic AI tasks. This new model offers up to 4x faster throughput and 30% faster task comp…
-
Speculative Decoding Matures, Accelerating LLM Inference
Speculative decoding, a technique for accelerating LLM inference, has matured significantly, with frameworks adopting it and users reporting impressive performance gains. While the core concept has existed for years, it…
-
Hugging Face highlights new inference providers and AI tools · 4 sources tracked
Hugging Face is highlighting several companies and projects that are enhancing its inference capabilities. DeepInfra has been featured as an inference provider, while Hcompany's HoloTab is introduced as an AI browser pa…
-
Hugging Face integrates Baseten for expanded serverless AI model access
Hugging Face has integrated Baseten as a new Inference Provider on its Hub, allowing developers to access a wider range of AI models through serverless inference. This collaboration enables seamless integration of model…
-
Self-hosting open-source speech-to-text models incurs hidden costs
Self-hosting open-source speech-to-text models like Whisper Large V3, Qwen3 ASR, and NVIDIA's Parakeet and Canary can appear free initially, but the total cost of ownership is significant. Beyond the model weights, user…
-
AssemblyAI: Self-hosting AI models costs more than managed APIs
AssemblyAI argues that while self-hosting open-source speech models like Whisper or Qwen3-ASR on platforms such as Baseten, Modal, or Fireworks may seem cost-effective on paper, the total cost of ownership is often high…
-
MiniMax AI to host open-source event with Moonshot, Baseten, Modal
MiniMax AI is co-hosting an open-source event on August 6th, featuring discussions with other builders in the field. The event will include a community demo spot. Other participating companies include Moonshot, Baseten,…
-
GLM-5.2 model updated with vision capabilities on Hugging Face
A new version of the GLM-5.2 large language model, now with vision capabilities, has been released on Hugging Face. This update integrates the vision encoder from the Kimi 2.6 model, addressing a previous limitation of …
-
Together integrates Kimi K3 into Cursor AI assistant
Together has partnered with Cursor to integrate Kimi K3 into the Cursor AI assistant. This integration allows users to access Kimi K3 via US-based inference infrastructure provided by Fireworks, Together, and Baseten, w…
-
Cursor IDE integrates Kimi K3 language model
The AI-powered IDE Cursor has integrated Kimi K3, a language model that performs comparably to frontier models on the CursorBench benchmark. This integration is available through US-based inference partners, including F…
-
Anthropic launches Claude Opus 5, OpenAI model breaches Hugging Face · 1 source tracked
Anthropic has released Claude Opus 5, a more efficient model that approaches the capabilities of Claude Fable 5 at half the price. This new model has reportedly outperformed on several coding and knowledge-work benchmar…
-
Google, OpenAI join 50-org push for open AI models; Anthropic absent
An open letter advocating for expanded compute access and avoiding premature restrictions on open-weight AI models has seen its signatory list double to 50 organizations, including major players like Google and OpenAI. …
-
Wuwenxiongqiong pivots AI infra strategy to tackle 'token assassins'
Wuwenxiongqiong, an AI infrastructure company, is evolving its strategy from connecting models and chips to a "front store, back factory, one center" Agentic Infra model. This shift addresses the increasing costs and in…
-
AI infrastructure firms raise $3.8B in four weeks for open-weight model services
Several companies specializing in managed infrastructure for open-weight AI models have recently secured substantial funding. Fireworks announced a $1.505 billion Series D, Baseten closed around $1.5 billion, and Togeth…
-
Fireworks AI sees surging demand for open-weight models
Fireworks AI has noted a significant increase in demand for open-weight inference and training, with their models appearing on both trending and fastest-growing lists. This surge is tracked by Ramp, which uses a databas…
-
French startup ZML releases free AI inference software for diverse chips · 2 sources tracked
French AI startup ZML has launched ZML/LLMD, a new software designed to optimize AI inference speed across a variety of hardware, including chips from Nvidia, AMD, Google, Apple, and Intel. The tool aims to break down v…
-
AI agents gain ML infrastructure control via Model Context Protocol
The Model Context Protocol (MCP) is enabling AI agents to manage machine learning infrastructure by moving beyond simple text prompting to structured execution. This protocol allows agents, like those integrated with Ba…
-
Nvidia shifts to vendor financing for AI cloud GPU deployments
Nvidia has launched a new financing program that allows AI cloud providers to acquire its GPUs without upfront capital expenditure. This initiative converts Nvidia's previous ad hoc customer financing into a repeatable …
-
NVIDIA partners with AI clouds to scale compute infrastructure
NVIDIA is launching a new business model to provide AI companies with large-scale compute infrastructure. This initiative aims to address the capital-intensive nature of AI factories by partnering with AI clouds and off…