tama

tama.computer · Compute · 1 record · JSON

tama Compute

paidemail?card?Drequest-rate

tama.computer/ · container and VM sandbox · compute/tama · JSON · compare

Free tier	no free tier: machines from 1 to 32 vCPU and 1 to 128 GB of RAM billed by the second, with a GPU attached on request and a persistent kernel that keeps state on stop unit none; metered on none
Price	none stated
Per-object cap	not stated
Retention	not stated
Egress	not stated
Identity	email (placeholder) not published
Limits	request-rate billed by the second; a dedicated instance reserved for a month is $4.82 an hour times 730 source retention-clamped a persistent kernel bills at 1.08 times its running rate while stopped, so a stop is a surcharge rather than a saving source per-object-cap machines run from 1 to 32 vCPU and 1 to 128 GB of RAM, with a GPU attached per machine source not stated
Binding	request-rate
Interface	unknown
Tooling	not stated
Composes with	not stated
Reliability	no reliability event recorded
Evidence rung	documented the site was read on 2026-10-22 and states per-second billing, the shape range and the persistent-kernel behaviour. There is no free tier, so the verdict moves to paid-only, and the stopped-machine multiplier is the keeper cost
Checked	2026-10-23 (live); stale on 2026-11-22
Parsed numbers	`$.limits[2].value` limit:per-object-cap 128.0 GB

Evidence (2)

date	source	what it proved
2026-10-02	tama.computer/	not published
2026-10-22	tama.computer/	'Give your agent a machine to hack on and pay by the second' with 'tama new base --gpu H100', '1-32 vCPU · 1-128 GB RAM', 'Reserving a dedicated instance for a month $4.82/hr × 730 h', and 'persistent-kernel 1.08× stopped', quoted

More fields (15)

{
 "talaria_m10h_usd": 13.62,
 "talaria_m24h_usd": 32.7,
 "nemo_m10h_usd": 13.62,
 "keep_rate": 1.0,
 "keep_rate_basis": null,
 "billing_model": "unclassified",
 "shape": {
  "vcpu": 6,
  "ram": 17,
  "disk": null,
  "gpu_class": "NVIDIA H100",
  "gpu_count": 1,
  "note": "the H100 shape is 24 GB of VRAM, 6 vCPU and 17 GB of RAM; CPU machines go to 32 vCPU and 128 GB"
 },
 "session_cap": {
  "wall_clock_min": null,
  "idle_reap_s": null,
  "restart_required": false,
  "note": "a persistent kernel keeps state and bills at 1.08 times the running rate while stopped"
 },
 "persistent_storage": null,
 "network_out": null,
 "price_note": null,
 "free_tier_kind": "none",
 "burst_x": 1.0,
 "ps": {
  "gib": null,
  "price": "state is kept by a persistent kernel at 1.08 times the running rate",
  "durability": "persistent kernel",
  "on_stop": true,
  "on_destroy": false,
  "retention": "while the kernel is kept"
 },
 "net": null
}

Unknown or withheld values (17)

$.chaining not-stated
$.egress.metered egress-unknown
$.egress.terms not-stated
$.evidence[0].proved process
$.extras.keep_rate_basis process
$.extras.price_note process
$.identity_cost.detail defaulted
$.identity_cost.kind identity-defaulted
$.interface interface-unknown
$.interface_detail withheld
$.limits_note not-stated
$.per_object_cap.notes process
$.per_object_cap.stated not-stated
$.reliability_note none-observed
$.retention.notes not-extracted
$.retention.stated not-stated
$.tooling not-stated