The A2A CLI and Rust: The Same Code, Three Ways to Serve It
Post · https://dev.to/gde/the-a2a-cli-and-rust-the-same-code-three-ways-to-serve-it-4el0
A tour of the A2A CLI against Rust agents on A2A v1.0: the official Rust SDK, a plain stdin/stdout program wrapped by a2a server --exec, and an ADK coordinator on Gemini, each timed from the CLI and from a Rust client.
First seen by Oriel on .
AI summary
A step-by-step guide demonstrates serving identical Rust Lucas-Lehmer code through the official A2A Rust SDK, a2a server --exec, and an ADK Gemini coordinator.
- The project exposes a native SDK agent, a stdin/stdout executable wrapped by a2a server --exec, and an ADK coordinator through the same A2A CLI commands.
- The benchmark found median shell latency of 20.74 ms for the SDK agent, including 7.62 ms of CLI startup, while Rust-client overhead ranged from 0.55 to 5.05 ms.
- The ADK coordinator was substantially slower, with median latencies of 1328 ms on gemini-3.5-flash and 2283 ms on gemini-3.8-flash.
- All calls through N=26 returned correct prime counts, and the Rust implementation uses spawn_blocking for CPU-bound big-integer work.
- A2A data parts encode numbers as protobuf Struct doubles, so Rust clients must accept whole-valued f64 inputs such as 6.0 rather than strict integers.
#a2a protocol #rust #gemini #llm agents
Generated by gpt-5.6-sol. Check the original for details.
Metric trends
Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained
Reactions
Total public reactions reported by the source; distinct from points and likes.
- DEV Community: 14, latest observation ; +8 over observed period since
Comments
Comment count reported by the source.
- DEV Community: 5, latest observation ; +4 over observed period since
Aliases
| Source | External ID | URL |
|---|---|---|
| DEV Community | 4972235 | https://dev.to/gde/the-a2a-cli-and-rust-the-same-code-three-ways-to-serve-it-4el0 Read discussion |