dev.llmlatency/llm-latency-tracker
LLM Latency Tracker
dev.llmlatency/llm-latency-tracker costs 170 tokens of context per session. It has 2 tool definitions. That is 0.1 % of a 200k window. Measured 2026-10-03.
Measured latency, time to first token and uptime for ~45 AI inference APIs, by region.
alive probed 2026-10-03 · HTTP 200
2tools
170tokens for the tool definitions
<0.1 %of a 200k context window
366 msto initialize
132 msto list tools
- Endpoint
- https://llmlatency.dev/mcp (streamable-http)
- Host
- llmlatency.dev
- Protocol
- 2025-06-18
- Reports itself as
- llmlatency-mcp 1.0.0
- Capabilities
- tools
- Instructions
- none
- Repository
- https://github.com/mazamaka/llm-latency-tracker
- Website
- https://llmlatency.dev
- Registry
- active, updated 2026-08-16 · registry entry
- Reachable
- 2 of the last 2 probes
Most expensive tools
| Tool | Tokens | Share of this server |
|---|---|---|
| get_model_deprecations | 93 | 55 % |
| get_ai_api_latency | 77 | 45 % |
History
We add a row when the result changes. A server with no change keeps its last row.
| Probed | Result | Init | Tools | Tokens |
|---|---|---|---|---|
| 2026-10-03 20:33 | alive | 366 ms | 2 | 170 |
| 2026-10-03 17:50 | alive | 603 ms | 2 | 170 |
Badge
[](https://mcp-pulse.ulehla.dev/s/dev.llmlatency/llm-latency-tracker)