mirror of
https://github.com/hanzoai/site.git
synced 2026-11-21 01:07:08 +00:00
platform.hanzo.ai is the platform console (with cloud.hanzo.ai redirecting to it at the edge). Point the remaining cloud.hanzo.ai links on /benchmarks and /philosophy directly at https://platform.hanzo.ai. Claude-Session: https://claude.ai/code/session_015XYKLix9PSyn6Jw2HWKprk
175 lines
4.2 KiB
JSON
175 lines
4.2 KiB
JSON
[
|
||
[
|
||
{
|
||
"p": "Benchmarks / Fleet residency"
|
||
},
|
||
{
|
||
"a": "Benchmarks",
|
||
"href": "/benchmarks"
|
||
},
|
||
{
|
||
"h1": "What a dormant agent costs"
|
||
},
|
||
{
|
||
"p": "A million dormant agents written to disk and read back: 477 bytes of state each, 455 MB for the million, and a resume in 0.034 ms. Measured on a laptop, against a competitor figure that was modelled and never billed."
|
||
},
|
||
{
|
||
"a": "hanzoai/benchmarks · fleet",
|
||
"href": "https://github.com/hanzoai/benchmarks/tree/main/fleet"
|
||
},
|
||
{
|
||
"a": "What a Dormant Agent Costs",
|
||
"href": "https://github.com/hanzoai/papers/blob/main/site/public/pdfs/hanzo-dormant-agents.pdf"
|
||
},
|
||
{
|
||
"code": "node fleet/fleet.mjs /tmp/fleet"
|
||
},
|
||
{
|
||
"p": "prints the table on this page — transcribed from bench/README.md, the pass of 2026-09-07"
|
||
},
|
||
{
|
||
"h2": "What is measured"
|
||
},
|
||
{
|
||
"p": "A million agents are created, written to disk, and read back. What is counted is the state a dormant agent occupies — the row that has to exist for an agent to be resumable at all — the rate at which those rows can be written, and the time to bring one back. Nothing is running while the measurement is taken, which is the point: an agent platform’s floor is what it costs to keep an agent that is doing nothing."
|
||
},
|
||
{
|
||
"h2": "Results"
|
||
},
|
||
{
|
||
"p": "one million agents · an M-series laptop · the pass of 2026-09-07"
|
||
},
|
||
{
|
||
"table": [
|
||
[
|
||
"measurement",
|
||
"Hanzo, measured",
|
||
"the published figure"
|
||
],
|
||
[
|
||
"state per dormant agent",
|
||
"477 bytes",
|
||
"~1 MB, a competitor’s own modelled assumption"
|
||
],
|
||
[
|
||
"one million dormant agents on disk",
|
||
"455 MB",
|
||
"~977 GB at that assumption"
|
||
],
|
||
[
|
||
"dormant agents written per second",
|
||
"309,789",
|
||
"—"
|
||
],
|
||
[
|
||
"resume one, in process",
|
||
"0.034 ms",
|
||
"—"
|
||
]
|
||
]
|
||
},
|
||
{
|
||
"p": "The right-hand column is not a competitor’s measurement. It is the per-agent figure a competing platform assumes in its own published arithmetic, marked there as modelled and never billed. Ours is a directory you can list. The distance between the two is about 2,200× and most of it is the difference between an assumption and a file."
|
||
},
|
||
{
|
||
"h2": "What this does not measure"
|
||
},
|
||
{
|
||
"p": "A dormant agent is not a running one. This number is the floor, and the floor is the easy part. What a live agent costs is a different measurement with a different command — 601 bytes of heap and 125 ns to spawn ."
|
||
},
|
||
{
|
||
"a": "601 bytes of heap and 125 ns to spawn",
|
||
"href": "/benchmarks/goroutine"
|
||
},
|
||
{
|
||
"p": "One machine, one day. This is a single pass on one laptop. It is reproducible by one command and it is not a fleet of machines under load, so read it as a floor that has been established rather than as a service level."
|
||
},
|
||
{
|
||
"p": "Hanzo AI Cloud"
|
||
},
|
||
{
|
||
"h3": "Build on Hanzo Cloud"
|
||
},
|
||
{
|
||
"a": "Start on Hanzo Cloud",
|
||
"href": "https://platform.hanzo.ai"
|
||
},
|
||
{
|
||
"a": "View Models & Pricing",
|
||
"href": "/pricing"
|
||
},
|
||
{
|
||
"a": "Previous LoCoMo-Conv",
|
||
"href": "/benchmarks/locomo-conv"
|
||
},
|
||
{
|
||
"a": "All Benchmarks Board",
|
||
"href": "/benchmarks"
|
||
},
|
||
{
|
||
"a": "Next Live agent footprint",
|
||
"href": "/benchmarks/goroutine"
|
||
},
|
||
{
|
||
"h3": "Related in runtime"
|
||
},
|
||
{
|
||
"a": "Live agent footprint",
|
||
"href": "/benchmarks/goroutine"
|
||
},
|
||
{
|
||
"a": "Sandbox cold start",
|
||
"href": "/benchmarks/sandbox"
|
||
},
|
||
{
|
||
"p": "LoCoMo · MemoryAgentBench · LongMemEval · CronQuestions · RepoBench-R · LoCoMo · subject scope · LoCoMo-Conv · Engine prefill on CUDA vs llama.cpp · GPQA-Diamond"
|
||
},
|
||
{
|
||
"a": "LoCoMo",
|
||
"href": "/benchmarks/locomo"
|
||
},
|
||
{
|
||
"a": "MemoryAgentBench",
|
||
"href": "/benchmarks/memoryagentbench"
|
||
},
|
||
{
|
||
"a": "LongMemEval",
|
||
"href": "/benchmarks/longmemeval"
|
||
},
|
||
{
|
||
"a": "CronQuestions",
|
||
"href": "/benchmarks/cronquestions"
|
||
},
|
||
{
|
||
"a": "RepoBench-R",
|
||
"href": "/benchmarks/repobench-r"
|
||
},
|
||
{
|
||
"a": "LoCoMo · subject scope",
|
||
"href": "/benchmarks/locomo-subject-scope"
|
||
},
|
||
{
|
||
"a": "LoCoMo-Conv",
|
||
"href": "/benchmarks/locomo-conv"
|
||
},
|
||
{
|
||
"a": "Engine prefill on CUDA vs llama.cpp",
|
||
"href": "/benchmarks/inference"
|
||
},
|
||
{
|
||
"a": "GPQA-Diamond",
|
||
"href": "/benchmarks/gpqa"
|
||
},
|
||
{
|
||
"p": "harness and the committed table: hanzoai/benchmarks · fleet · the paper: What a Dormant Agent Costs"
|
||
},
|
||
{
|
||
"a": "hanzoai/benchmarks · fleet",
|
||
"href": "https://github.com/hanzoai/benchmarks/tree/main/fleet"
|
||
},
|
||
{
|
||
"a": "What a Dormant Agent Costs",
|
||
"href": "https://github.com/hanzoai/papers/blob/main/site/public/pdfs/hanzo-dormant-agents.pdf"
|
||
}
|
||
]
|
||
] |