The independent benchmark for compute
- Tweets 269
- Following 94
- Followers 1,314
- Likes 414
Pinned Tweet
🚨 Benchmark Friday 🚨
After weeks of work from us and the participants, today we are excited to release the results of the 2026 100k Scale Invitational.
Over the next hour we'll be releasing each set of results.
Huge shoutout to @namespacelabs for powering our benchmarks!
5
2
47
627,387
Another day and another adaptor.
Except this one has NATIVE support for snapshots, forks, etc plus git semantics.
We've just added in support for @pierrecomputer's code storage to the StorageSDK.
1
5
5,289
Storage SDK now has support for @archildata.
With native snapshotting and forking coming soon!
1
1
7
6,618
ComputeSDK retweeted
The latest browser benchmark isn't just “how fast can you launch a browser?” but “once the browser is alive, how fast can an agent actually use it?”
50 sequential actions per session. Navigate, wait, screenshot, read, click, go back, repeat.
🚨Benchmark Friday🚨
This friday we have a NEW browser benchmark dropping.
It's focused on agent ACTIONS.
see thread for more details. 🧵
2
1
5
1,087
🚨Benchmark Friday🚨
This friday we have a NEW browser benchmark dropping.
It's focused on agent ACTIONS.
see thread for more details. 🧵
2
3
1,310
They use very similar tech isolation, microvms typically, etc.
But the primary interesting thing is actions taken by agents.
1
83
If intereted in the new benchmark take a look at this PR. we'd love feedback!
github.com/computesdk/benchm…
69
ComputeSDK retweeted
We also did so many test runs!
This whole benchmark stretched @computesdk quite a lot and it was invaluable to have @namespacelabs in the weeds with us.
We are happy to support @computesdk!
They launched thousands of instances, each one issuing thousands of concurrent requests, on a software stack that was continuously being updated.
Exactly the type of workloads Namespace is designed for.
1
1
8
1,226
ComputeSDK retweeted
We are happy to support @computesdk!
They launched thousands of instances, each one issuing thousands of concurrent requests, on a software stack that was continuously being updated.
Exactly the type of workloads Namespace is designed for.
Last week we announced the results of the scale invitational. Link to the full results in the 🧵
It's our latest benchmark powered by @namespacelabs.
With backing from our excellent long term partners.
@latitudesh
@browserbase
@TigrisData
@neondatabase
@GitBookIO
2
1
16
3,939
Last week we announced the results of the scale invitational. Link to the full results in the 🧵
It's our latest benchmark powered by @namespacelabs.
With backing from our excellent long term partners.
@latitudesh
@browserbase
@TigrisData
@neondatabase
@GitBookIO
3
1
19
312,652
You can find the full results here: platform.computesdk.com/scal…
2
131
More requests for 1m sandboxes!
What's our secret? That this is business as usual for us, and has been for a long time.
We have many customers running 100k or even 1M sandboxes with us on a regular basis. Large RL runs for instance.
Looking forward to the 1M sandbox challenge – we are ready any time!
2
1,503
🚨🚨🚨
Huge congrats to @isorun for completing the 100k Scale Invitational.
See the full results below:
2
2
2
1,161