Submodule bindings for vllm.utils.
Version
- Requested: 0.14.0
- Observed at generation: 0.14.0
Runtime Options
All functions accept a __runtime__ option for controlling execution behavior:
Vllm.Utils.some_function(args, __runtime__: [timeout: 120_000])Supported runtime options
:timeout- Call timeout in milliseconds (default: 120,000ms / 2 minutes):timeout_profile- Use a named profile (:default,:ml_inference,:batch_job,:streaming):stream_timeout- Timeout for streaming operations (default: 1,800,000ms / 30 minutes):session_id- Override the session ID for this call:pool_name- Target a specific Snakepit pool (multi-pool setups):affinity- Override session affinity (:hint,:strict_queue,:strict_fail_fast)
Timeout Profiles
:default- 2 minute timeout for regular calls:ml_inference- 10 minute timeout for ML/LLM workloads:batch_job- Unlimited timeout for long-running jobs:streaming- 2 minute timeout, 30 minute stream_timeout
Example with timeout override
# For a long-running ML inference call
Vllm.Utils.predict(data, __runtime__: [timeout_profile: :ml_inference])
# Or explicit timeout
Vllm.Utils.predict(data, __runtime__: [timeout: 600_000])
# Route to a pool and enforce strict affinity
Vllm.Utils.predict(data, __runtime__: [pool_name: :strict_pool, affinity: :strict_queue])See SnakeBridge.Defaults for global timeout configuration.
Summary
Functions
Calculate the request length (in number of tokens) give either
Python module attribute vllm.utils.MASK_64_BITS.
Python binding for vllm.utils.random_uuid.
Functions
@spec length_from_prompt_token_ids_or_embeds(term(), term(), keyword()) :: {:ok, integer()} | {:error, Snakepit.Error.t()}
Calculate the request length (in number of tokens) give either
prompt_token_ids or prompt_embeds.
Parameters
prompt_token_ids(term())prompt_embeds(term())
Returns
integer()
@spec mask_64_bits() :: {:ok, integer()} | {:error, Snakepit.Error.t()}
Python module attribute vllm.utils.MASK_64_BITS.
Returns
integer()
@spec random_uuid(keyword()) :: {:ok, String.t()} | {:error, Snakepit.Error.t()}
Python binding for vllm.utils.random_uuid.
Returns
String.t()