#inference
Inference: 8 AI articles covering inference news, analysis, and research
Articles
Fractile Raises $720M to Challenge Nvidia in the AI Chip Raceβ8
UK AI chip startup Fractile has raised $720M in Series C funding to challenge Nvidia with a full-stack, DRAM-based chip design aimed at cutting AI inference cos...
Handling Variable-Length Prompts on Chimera GPNPU: A 2026 Perspectiveβ8
Discover how the Chimera GPNPU handles variable-length prompts in 2026 with adaptive scheduling, elastic buffering, and fine-grained parallelism for efficient o...
How to Define CPU and Software Stack Performance for Agentic AIβ9
Agentic AI shifts performance to CPU orchestration, concurrency, and software stack efficiency. Learn the key metrics that define end-to-end agentic workload pe...
Etched Fielding Funding Offers at $40B+ Valuation as AI Chipβ9
Etched is in talks to raise new funding at a $40B+ valuation as AI chip demand surges, doubling its last round amid Nvidia competition.
NVIDIA Unveils DGX Spark 64GB: A 1-PetaFLOP Grace Blackwellβ8
NVIDIA's DGX Spark 64GB delivers 1 PetaFLOP Grace Blackwell performance for local AI agents, inference, and fine-tuningβscalable to 128GB by clustering two unit...
Turning Edge AI Data Into Real-Time Actionβ9
Edge AI moves from concept to production in 2026. Explore TinyML, hardware acceleration, and real-time pipelines turning edge data into instant action.
GGUF vs GPTQ vs AWQ vs EXL2: LLM Model Formats Explained (2026)β10
GGUF vs GPTQ vs AWQ vs EXL2 explained for 2026: compare LLM model formats, containers, and quantization methods to choose the right one for deployment.
Etched's Valuation Doubles to $21B in a Month, Fueled by Janeβ7
Etched's valuation soars to $21B after Jane Street-led funding, fueled by new low-voltage inference chips and cluster-scale memory technology.
