Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.inference.net/llms.txt

Use this file to discover all available pages before exploring further.

Inspect inference requests and responses captured by Observe. List recent inferences in the active project, filter by model or task id, and drill into the full request and response payloads. Alias: inf inferences

​
inf inference list

Display recent inferences in the active project. 
inf inference list
Alias: inf inference ls

​
Options

FlagRequiredDescriptionDefault
-l, --limit <n>NoMaximum number of results20
-m, --model <model>NoFilter by model nameAll models
--task <taskId>NoFilter by task ID — matches the x-inference-task-id header set by inf instrument --task-id—
--cursor <cursor>NoPagination cursor (from a previous response)—
The table shows the inference ID (8-char prefix), model, status code (color-coded), input/output token counts, latency in ms, cost, and timestamp. Use --json to get full UUIDs for scripting. When more results are available, the CLI prints a pagination cursor you can pass back with --cursor.

​
Examples

# List the 10 most recent inferences
inf inference list --limit 10

# Filter by model
inf inference list --model meta-llama/llama-3.1-8b-instruct/fp-8

# Filter by the task id the instrumented call tagged itself with
inf inference list --task support-tickets

# Page forward using the cursor printed by the previous call
inf inference list --cursor eyJpZCI6...

​
inf inference get

View the full request and response for a specific inference — method, path, request body, and response body all printed as formatted JSON. Useful for debugging specific API calls or inspecting model behavior. 
inf inference get <id>

​
Arguments

ArgumentRequiredDescription
idYesFull inference UUID or a 4+ character prefix

​
Examples

# Using a 4+ character UUID prefix
inf inference get inf_abc1

# Using a full UUID from --json output
inf inference list --json | jq -r '.items[0].id' | xargs inf inference get