GPU Taskman shows per-process VRAM and an LLM weight and KV split
MIT-licensed GPU Taskman v0.3.3, released 7 November 2026, reports per-process VRAM and estimates LLM weights and KV cache.
Topic · 12 stories
12 stories tagged gpu on AI Tamer, newest first.
MIT-licensed GPU Taskman v0.3.3, released 7 November 2026, reports per-process VRAM and estimates LLM weights and KV cache.
cuda-oxide 0.3.0 drops the pinned nightly, shares one host runtime with cutile-rs, and previews CuTe in Rust. What changed and what is still missing.
Open Linux modules lift GeForce FP16 and BF16 tensor issue rates toward 2x on RTX 30 and 40 cards. Owners report 14 to 36% faster LLM prefill, not decode.
doubleAI's Maxwell optimizes programs while they run and says it formally proves each rewrite equivalent. It launched with a $200M Series A disclosure.
At OCP, Intel named Crescent Island the GPU D-series. GPU D270 is a 350 W air-cooled PCIe card. Specs are Intel's. No price or ship date.
MIG cuts a supported NVIDIA GPU into up to seven instances with their own memory, cache and compute. How to create them and pick a profile for a model server.
NVIDIA's bulletin rates CVE-2026-47483 at 8.2: unauthenticated pprof requests can exhaust DCGM Exporter. Lava found about 2,100 hosts exposing exporter metrics. NVIDIA lists 4.8.2 as the fix.
Continuous batching lets a serving system replace finished requests while longer responses keep generating. Here is how that changes GPU use and what limits it.
Device plugins advertise GPUs as countable resources. That changes how Pods request them and how many workloads a node can schedule.
Vectors, Plainly · Part 6 of 6
The parts tutorials skip: sharding and write freshness, memory and disk-based indexes, where a GPU actually pays off, and the security and compliance questions a vector store raises.
Modal (Oct 1, 2026) made Clusters generally available via @modal.clustered for multi-node GPU jobs. RDMA is optional; Volumes, Cloud Bucket Mounts, and Queues stay in the loop. No dollar rates on the post.
A dated brief on NVIDIA’s Sep 2026 CUDA Rust post: two early-stage kernel tracks — cuda-oxide (SIMT, early alpha) and cutile-rs (Tile) — neither production-ready. Correctness demo only; no perf claims.