GPU Manager MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| GPU_MANAGER_DB_PATH | No | SQLite location for local history. | ~/.local/share/gpu-manager-mcp/history.sqlite3 |
| GPU_MANAGER_CACHE_TTL | No | Telemetry cache TTL in seconds for fast repeat queries. | 1.0 |
| GPU_MANAGER_HTTP_HOST | No | Loopback-only HTTP bind host. | 127.0.0.1 |
| GPU_MANAGER_HTTP_PORT | No | Loopback-only HTTP bind port. | 8940 |
| GPU_MANAGER_HTTP_ENABLED | No | Enable Streamable HTTP transport. | false |
| GPU_MANAGER_COMMAND_TIMEOUT | No | Timeout in seconds for allowlisted helper CLIs. | 10 |
| GPU_MANAGER_SAMPLE_INTERVAL | No | Background sampling period in seconds. | 2.0 |
| GPU_MANAGER_HISTORY_MAX_ROWS | No | Hard row cap for bounded local history storage. | 200000 |
| GPU_MANAGER_HISTORY_RETENTION_HOURS | No | History window kept in hours. | 24 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_gpusA | Discover all GPUs (NVIDIA/AMD/Intel). Returns stable gpu_ids, names, VRAM totals, PCI bus ids, driver versions and supported telemetry capabilities per device. |
| get_gpu_infoC | Hardware and driver information for one GPU. |
| get_system_gpu_summaryB | System-wide overview: every GPU with one-line key metrics. |
| get_gpu_utilizationC | GPU core utilization percentage for one GPU. |
| get_gpu_memoryC | VRAM total/used/free in bytes for one GPU. |
| get_gpu_temperatureB | GPU temperature in Celsius (edge sensor where available). |
| get_gpu_powerC | GPU package power draw in watts. |
| get_gpu_clocksC | Current graphics clock in MHz. |
| get_gpu_processesB | Processes using the GPU with per-process VRAM, when the driver exposes it (NVML or rocm-smi). Returns supported=false otherwise. |
| get_all_gpu_metricsB | Full telemetry snapshot for every GPU at once. |
| get_gpu_historyB | Historical samples for one metric over a time window. Metrics: utilization_percent, memory_used_bytes, memory_total_bytes, memory_free_bytes, temperature_celsius, power_watts, clock_mhz. |
| get_gpu_peak_usageC | Max/min/avg of a metric over a window. |
| get_gpu_healthB | Health check: thermal/memory/power issues from live + historical data. |
| estimate_model_fitB | Estimate whether an LLM fits in one GPU's VRAM. Rough estimate: weights + runtime overhead + KV cache. Pass 'architecture' (e.g. llama-3-8b, mistral-7b) for an architecture-aware KV cache; without it, default GQA assumptions are used and flagged. |
| recommend_gpu_for_modelC | Rank all discovered GPUs by estimated fit for the given model. |
| find_available_gpuA | GPUs with at least required_vram_gb (GiB) free right now. |
| compare_gpusB | Side-by-side comparison of specs and live telemetry for several GPUs. |
| detect_gpu_bottleneckB | Heuristic bottleneck analysis (compute/thermal/power/memory) from sampled history. Reports when evidence is insufficient. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 18 tools
The per-metric tools (utilization, memory, temperature, power, clocks) are clearly scoped to one signal each, and the aggregate tools (get_all_gpu_metrics, get_system_gpu_summary, compare_gpus) are distinguishable by their descriptions. Minor overlap remains between list_gpus, get_gpu_info, and get_system_gpu_summary, which all return device-level inventory/summary data that could confuse selection.
All 18 tools use snake_case verb_noun phrasing (get_gpu_memory, get_gpu_history, estimate_model_fit, recommend_gpu_for_model, find_available_gpu, detect_gpu_bottleneck). The get_gpu_* telemetry family is highly predictable, and the analysis tools follow the same verb-first convention.
18 tools is on the heavy side for a GPU monitoring server, and the narrow per-metric getters largely duplicate what get_all_gpu_metrics returns. Still, each tool has a single defensible purpose and the surface is far from bloated given the breadth of discovery, telemetry, history, and advisory operations.
Coverage is strong: discovery, per-metric and bulk telemetry, historical samples, health checks, model-fit estimation, and bottleneck analysis form a coherent lifecycle. Gaps are modest — no fan speed, ECC/memory-error counters, MIG/partition details, or any configuration/mutation operations, but agents can work around these for read-only monitoring.