Skip to main content
POST
GPU VRAM Calculator

Authorizations

Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json
ioModelPreset
enum<string>

Model Preset

Available options:
custom,
gpt2-124m,
llama-3-2-1b,
llama-3-2-3b,
mistral-7b,
llama-3-8b,
llama-2-13b,
mixtral-8x7b,
llama-3-70b,
llama-3-1-405b
ioParams
number

Parameters (B)

Required range: 0.001 <= x <= 2000
ioHiddenDim
number

Hidden Dim

Required range: 64 <= x <= 65536
ioLayers
number

Layers

Required range: 1 <= x <= 256
ioMode
enum<string>

Workload

Available options:
inference,
training
ioPrecision
enum<string>

Precision

Available options:
fp32,
fp16,
int8,
int4
ioBatchSize
number

Batch Size

Required range: 1 <= x <= 4096
ioSeqLen
number

Sequence Length

Required range: 1 <= x <= 1048576
ioOptimizer
enum<string>

Optimizer

Available options:
adam,
sgd-momentum,
sgd
ioMixedPrecision
boolean

Mixed precision (fp32 master weights + grads)

ioGradCheckpoint
boolean

Gradient checkpointing (recompute activations)

Response

Tool output

tool
string
required
tool_version
string
required

Output-contract version for this tool. Currently "1" for all tools.

outputs
object
required
credits_used
integer
required

Credits this call consumed, after any settlement refund. 0 when metering is disabled.

request_id
string
credits_remaining
integer | null

Credits left in the current monthly allowance, or null when metering is disabled.