Pi coding agent extension: shows current GPU VRAM usage in the footer (e.g. VRAM 10.8GB/16.3GB)

Otto Brain - self-contained host for local GGUF models, with measured VRAM budgeting and reasoning-budget control

OpenCode plugin to display live GPU VRAM usage, compute utilization and temperature in the session prompt.

DeepSeek Harness plugin: LLM model selection & deployment analysis (deployability, VRAM, TTFT, latency, throughput, power for 38 models × 20 GPUs/NPUs)

OpenCode plugin to display live GPU VRAM usage, compute utilization and temperature in the session prompt.

VRAM JSON cache with IndexedDB integration, built with TypeScript.

MCP server for the MakerPortal compute API — biquad filter design, rectangular-room eigenmodes, and LLM VRAM fit, computed by the same functions that render makerportal.ai.

Enterprise-grade VRAM management for HTML5 games. Features hardware tier detection, priority-based hysteresis eviction, Safari throttle detection, and safe scene streaming.

Model Context Protocol server for AI VRAM calculation and GPU recommendation

Floating GPU monitor inside the DSH Web UI: live per-GPU utilization, VRAM, temperature, power and top processes from nvidia-smi. · DSH Web 内置悬浮 GPU 监控：实时显示每张 GPU 的利用率、显存、温度、功耗与占用进程（nvidia-smi）。

CLI preflight checks for model VRAM, CUDA, GPU compatibility, and quantization risk.

Zero-GC WebGL2 render-path telemetry. Deterministic command counters (draw calls, instances, floats uploaded) recorded in the GL-agnostic core, plus an async EXT_disjoint_timer_query_webgl2 GPU-time pool. Passive instrument; pairs with lite-vram (the mana

MCP server that inspects local GGUF and safetensors model files — architecture, quantization, parameters, tensors, and VRAM-fit estimates from headers only, without loading the model

Can my machine run this model? Estimate VRAM/RAM, tok/s, and TTFT.

Detect your hardware and find LLMs that will actually run on it — with real VRAM/KV-cache fit math, not opaque scores.

GPU VRAM probing — memory usage tracking, allocation monitoring, OOM prevention

Freelens extension for GPUs — per-pod and per-device GPU utilisation, VRAM and power from dcgm-exporter or a per-process exporter, zero cluster footprint

Run local models in opencode. llama.cpp as a provider, with live GPU, VRAM and CPU telemetry in the sidebar.

Hardware-aware local-LLM loadout planner: chooses model class, quant, KV-cache, context, and offload policy for your VRAM/RAM, and refuses configs that would silently page to disk.

Zero-GC off-thread ImageBitmap loader with strict VRAM limits, URL deduplication, and LRU eviction to prevent mobile browser crashes.