GPU agent sandboxes: prices at 2 vCPU / 4 GB (Nov 2026)

GPU agent sandboxes: prices at 2 vCPU / 4 GB (Nov 2026)

#	Provider	1 h busy	1 h at 10% CPU	Fleet month	Pay	At once	Starts	Caveats
1	Koyebeco-large (2 vCPU / 4 GB) Pay as you go · fleet ≥ $253.44 Not published: minimum	≥ $0.0288[1]	≥ $0.0288	≥ $253.44	Pay as you go	?	?	Not published: minimum
2	Northflank2 vCPU / 4 GB Pay as you go · fleet ≥ $586.61 · 500 at once Not published: minimum	≥ $0.0667[2]	≥ $0.0667	≥ $586.61	Pay as you go	500	?	Not published: minimum
3	Daytona2 vCPU / 4 GB Pay as you go · fleet ≥ $1,457 · starts 300/min then refused Not published: minimum	≥ $0.166[3]	≥ $0.166	≥ $1,457	Pay as you go	?	300/minthen refused	Not published: minimum
4	Tensorlake2 vCPU / 4 GB Prepaid credit · fleet ≥ $1,760 · 100 at once Active-CPU billing Not published: minimum	≥ $0.2[4]	≥ $0.074	≥ $1,760	Prepaid credit	100	?	Active-CPU billing Not published: minimum
5	Beam2 vCPU / 4 GB Plan from $20/mo · fleet ≥ $2,088 · 30 at once Not published: minimum	≥ $0.227[5]	≥ $0.227	≥ $2,088on Team	Plan from $20/mo	30	?	Not published: minimum
6	Modal2 vCPU / 4 GB Pay as you go · fleet ≥ $2,094 · 100 at once Not published: minimum	≥ $0.238[6]	≥ $0.238	≥ $2,094	Pay as you go	100	?	Not published: minimum

Not published, so not listed here: Novita AI, Google Cloud Agent Platform sandboxes, Islo, Runloop.

1 h busy: all CPU billed. 1 h at 10% CPU: lower only where idle CPU isn't billed. Fleet month: 50 × 8 h × 22 days busy, on the lowest plan that runs 50 at once (its fee, less usage it includes). At once and Starts: on the terms priced. ≥ = a term isn't published. Method