Models & Labs
vLLM v0.26.0 Release Enhances Model Support
vLLM Releaseshigh confidence
Why it matters
- →The new Inkling model family introduces advanced features for improved AI model performance.
- →Enhanced hardware support broadens the applicability of vLLM across different platforms.
- →Improved attention mechanisms and KV offloading increase efficiency for hybrid models.
vLLM has released version 0.26.0, featuring significant updates including the new Inkling model family with advanced CUDA graph support and speculative decoding. The release also enhances performance across multiple hardware platforms, such as AMD and XPU, with the DeepSeek-V4 performance push. Improvements in attention mechanisms and KV offloading provide greater flexibility for hybrid models. These updates make vLLM a more powerful tool for developers handling large-scale AI models.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
How we got here
- 1llama.cpp b9272 Release Adds New Features
llama.cpp Releases · May 22, 2026 · Related
- 2llama.cpp b9330 release improves model performance
llama.cpp Releases · May 27, 2026 · Related
- 3vLLM v0.22.0 Release Enhances Model Performance
vLLM Releases · May 31, 2026 · Same story
- 4vLLM Boosts Transformers Backend for Native-Speed Inference
Hugging Face Blog · July 8, 2026 · Related
- 5vLLM v0.25.0 Release Enhances Model Execution
vLLM Releases · July 13, 2026 · Same story
What happened next
- 6Liquid AI releases DSpark drafter for vision models
Hugging Face Blog · September 24, 2026 · Related
More from vLLM Releases
More in Models & Labs
Models & Labsother
llama.cpp b12340 adds CUDA 13 and ROCm 10.0
This release quietly closes the hardware gap for local inference by adding default support for CUDA 13 and ROCm 10.0 across Linux and Windows. Users with newer NVIDIA GPUs or AMD accelerators no longer need to compile from source to get performance, while Apple Silicon builds now ship KleidiAI disabled by default, shifting the focus back to standard ARM optimizations. The inclusion of Snapdragon support for Linux and Android marks a significant step toward on-device AI beyond the x86/ARM desktop ecosystem. It is less about new features and more about ensuring llama.cpp remains the universal runtime for every available accelerator.
llama.cpp Releases
Models & Labscoding
llama.cpp b12342 adds few-row MMQ for speculative decoding
Speculative decoding just got a significant speed boost on NVIDIA GPUs. This release introduces a specialized kernel for 'few-row' matrix multiplications, targeting the small batch sizes typical of verification passes in speculative decoding workflows. By optimizing these specific operations for quantized formats like Q8_0 and IQ4 variants, llama.cpp reduces latency where it matters most during token generation. The change is CUDA-only for now but leverages hardware capabilities from Ampere onwards, making local inference faster without requiring new model architectures or complex configuration changes.
llama.cpp Releases
© TechCrunch AIModels & Labsmodels
Google deploys Gemini 4 Hydrogen for faster Search
Google is quietly shifting its search infrastructure by replacing older fast models with Gemini 4 Hydrogen, a tier designed specifically for speed and cost-efficiency rather than raw reasoning power. This model completes tasks 2.5 times faster than its predecessor while beating the more powerful Argon on specific intelligence benchmarks, proving that specialized efficiency can outperform brute force in high-volume workflows. By integrating this into Search's fan-out technique, Google now breaks down complex queries into subqueries with greater precision and speed, delivering richer visual responses without the latency of heavier models. This marks a strategic pivot toward agentic execution layers within consumer search, making AI Mode feel snappier for everyday multi-part questions.
TechCrunch AI