Targon · Rate Limits

Targon Rate Limits

Targon meters access to its OpenAI-compatible inference API per API key. Because inference is served by a decentralized marketplace of miners on Bittensor Subnet 4, effective throughput depends on available miner capacity for a given model in addition to any account-level request/token limits. Specific per-key or per-model limit values are not published in a single reference and are not reconciled in this artifact.

Targon Rate Limits is the machine-readable rate-limit profile for Targon on the APIs.io network, conforming to the API Commons Rate Limits specification.

It captures 3 rate-limit definitions, measuring requests, tokens, and throughput.

The profile also includes 2 backoff/retry policies defined and response codes documented for throttled.

Tagged areas include AI, LLM, Inference, Decentralized, and Bittensor.

3 Limits Throttle: 429
AILLMInferenceDecentralizedBittensorRate LimitingQuotasThrottling

Limits

Requests Per Minute (RPM) account
requests
see provider documentation
Per-key request rate; value not published, varies by account.
Tokens Per Minute (TPM) account
tokens
see provider documentation
Per-key token rate; value not published, varies by account and model.
Model Capacity network
throughput
depends on active miners
Effective throughput depends on miner capacity serving a given model on Subnet 4.

Policies

Decentralized Capacity
Throughput scales with the number of miners serving a model; capacity is market-driven rather than fixed by a single vendor.
Backoff Strategy
Clients should implement exponential backoff with jitter and honor Retry-After on 429 responses.

Sources