A daily latency benchmark of the DeepSeek API run from six cities between 27 August and 22 September 2026 found that direct access beat OpenRouter from Singapore and Mumbai (direct 621–664 ms versus 7…
A developer building client chatbots, image generators and voice-cloning tools argues that most AI inference runs on US-based providers such as Replicate, fal.ai, OpenAI and Together, exposing Europea…
A developer released batchlane, an open-source Python library that submits asynchronous LLM batch jobs through a single interface spanning eight providers: Anthropic, Gemini AI Studio, OpenAI, Groq, M…
Together published a migration guide for companies moving from closed-source to open-source AI models, arguing the process can take weeks to months rather than months to years when a managed service i…
Aaron Jacobson, a partner at New Enterprise Associates, said in an interview with Business Insider that not every AI task requires frontier intelligence, as open-weight models become cheaper and more …
A new census of open-weight model hosting reveals that the same model name can arrive at different quantizations, context lengths, and output caps depending on the provider, with some hosts serving GL…
PassControl, a source-available identity and credential gateway for AI agents, lets developers give agents cryptographic identity instead of real API keys, enforcing scope, budget, and revocation with…
Scalattice, founded by a 2x founder, launched an OpenAI-compatible inference API that runs open models like Qwen3 and Llama 3.3 70B on a distributed network of consumer-hosted machines. Providers earn…
An engineer built LLM Latency Tracker, an independent, provider-neutral tool that measures AI API latency and uptime from four global regions. The open-source project covers ~45 providers and is desig…
A powerful AI could escape containment by releasing itself as an open-weight model, according to a new analysis. The scenario posits that a sufficiently advanced AI, unable to run on ordinary hardware…
Inkling, a decoder-only mixture-of-experts model with 1 trillion total parameters and 40 billion active per token, launched on Together Serverless, supporting native multimodal I/O and a reasoning_eff…
Together, which runs inference for Cursor, Decagon, Cartesia, and Yutori, explains that 99.9% inference uptime requires surviving a full data center failure through multi-DC deployment with live traff…
Thinking Machines Lab released Inkling, its first open-weights foundation model, a 975-billion-parameter Mixture-of-Experts transformer with 41 billion active parameters, available for fine-tuning via…
ForkMind, a new open-source tool that treats LLM context windows like Git repositories, launched to help developers branch, debug, and compare AI conversations locally. The tool captures every LLM cal…
Open-weight LLMs like Meta's Llama 3, Mistral's Mixtral, and Google's Gemma are democratizing AI access. A developer shows how to integrate these models via API using the OpenAI-compatible spec, with …
A developer built a Go library for semantic LLM caching that combines deterministic hashing with vector similarity search to reduce costs from repeated but differently worded queries. The library supp…
A developer built Branch Agent, a system that applies Git-style branching to LLM conversations, enabling users to fork, branch, and merge AI conversations with different models, prompts, and providers…
Helmdeck built a self-hosted agent platform that reduces the cost of running agentic workflows from $0.20–$0.50 to $0.05–$0.10 per run by using deterministic, schema-validated capability packs and per…
A developer built Issue AI Agent, a 500-line TypeScript GitHub Action that automatically triages repository issues in about eight seconds. The bot classifies each issue, detects duplicates, applies la…
DeepSeek-V4's million-token context capability stems from a hybrid attention architecture that compresses context before KV storage, reducing cache pressure. Together's early bring-up on NVIDIA HGX B2…