w3-io/W3 Hyperbolic Action

AI inference and GPU compute via Hyperbolic — text, image, audio, vision, and on-demand GPUs

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
commandOperation: chat, generate-image, generate-audio, analyze-image, rent-gpu, list-gpus, terminate-gpuyes
api-keyHyperbolic API keyyes
api-urlAPI base URL overridenohttps://api.hyperbolic.xyz
modelModel ID (e.g. "deepseek-ai/DeepSeek-V3", "meta-llama/Llama-3.3-70B-Instruct")no
messagesChat messages as JSON array (e.g. [{"role":"user","content":"Hello"}])no
temperatureSampling temperature (0.0-2.0)no
top-pNucleus sampling threshold (0.0-1.0)no
max-tokensMaximum tokens to generateno
response-formatResponse format as JSON (e.g. {"type":"json_schema","json_schema":{...}})no
seedRandom seed for reproducibilityno
stopComma-separated stop sequencesno
toolsTool definitions as JSON array for function callingno
promptText prompt for image generation or question for vision analysisno
heightImage height in pixels (default: 1024)no
widthImage width in pixels (default: 1024)no
stepsNumber of diffusion steps (default: 25)no
loraLoRA style adapter name (e.g. "Pixel_Art", "Logo", "Sci-fi")no
lora-weightLoRA influence weight (0.0-1.0, default: 0.8)no
textText to convert to speechno
languageTTS language: EN, ES, FR, ZH, JP, KR (default: EN)no
speakerTTS speaker/accent (e.g. "EN-US", "EN-BR", "EN-AU")no
speedTTS speed multiplier (0.1-5.0, default: 1.0)no
image-urlURL of image to analyzeno
image-base64Base64-encoded image to analyzeno
gpu-typeGPU type for rental (e.g. "H100-SXM", "A100-SXM", "RTX-4090")no
gpu-countNumber of GPUs (1, 2, 4, or 8)no1
instance-idInstance ID for terminate-gpuno
namedescription
resultJSON result of the operation