🐛 Bug: Hardcoded Magic-Number Timeout in deploy_model()

File: src/opengradient/client/alpha.py — deploy_model()

What's broken

Every wait_for_transaction_receipt call in this file consistently uses the defined constants INFERENCE_TX_TIMEOUT = 120 or REGULAR_TX_TIMEOUT = 30, except for one outlier in deploy_model() (line 318) which uses a hardcoded 60:

# deploy_model() — line 318
tx_receipt = self._blockchain.eth.wait_for_transaction_receipt(tx_hash, timeout=60)  # ← magic number!

Contract deployment is the most resource-intensive transaction type (deploys full bytecode on-chain). It should use INFERENCE_TX_TIMEOUT = 120 at minimum, not an arbitrary 60s that:

1. Is inconsistent with every other call in the same file
2. Can cause premature TimeExhausted failures on congested networks
3. Provides no single place to tune the timeout if needed

Fix

tx_receipt = self._blockchain.eth.wait_for_transaction_receipt(tx_hash, timeout=INFERENCE_TX_TIMEOUT)

Impact

Files Changed

• src/opengradient/client/alpha.py