first_token
Method
This is a method of the StreamingTracker class in the profiler module.Mark time to first token.
Signature
Source
View on GitHub
praisonai/profiler.py at line 167Guide
Reference
- execute
- Execute Once
- Get Stats
- start
- stop
- Generate Crew And Kickoff
- Is Function Or Decorated
- Load Tools From Module
- Load Tools From Module Class
- Load Tools From Package
- Load Tools From Tools Py
- Convert And Save
- Discover Tools For Topic
- generate
- Get User Content
- Merge With Existing Agents
- Recommend Pattern
- Analyze Complexity
- Get Available Tools
- Create API File
- Create Dockerfile
- Run Commands
- Set Environment Variables
- deploy
- deploy
- Deploy Once
- Set Deployer
- start
- stop
- Get Imports
- Get Slowest
- execute
- API Call
- block
- clear
- cprofile
- disable
- enable
- Export Flamegraph
- Export Html
- Export Json
- Export To File
- Get API Calls
- Get Cprofile Stats
- Get Files Accessed
- Get Flamegraph Data
- Get Flow
- Get Imports
- Get Line Profile Data
- Get Memory Records
- Get Statistics
- Get Streaming Records
- Get Summary
- Get Timings
- Is Enabled
- memory
- Memory Snapshot
- Record API Call
- Record Flow
- Record Import
- Record Memory
- Record Streaming
- Record Timing
- report
- Set Line Profile Data
- streaming
- Streaming Async
- parse
- chunk
- Elapsed Ms
- end
- First Token
- start
- Ttft Ms
- Check Gpu
- Check Ram
- Create And Push Ollama Model
- inference
- Load Config
- Load Datasets
- Load Model
- Prepare Model
- Prepare Modelfile Content
- Print System Info
- Process Dataset
- Push Model Gguf
- run
- Save Model Gguf
- Save Model Merged
- Tokenize Dataset
- Train Model
- Check Gpu
- Check Ram
- Convert Sample
- Create And Push Ollama Model
- Load Config
- Load Datasets
- Prepare Model
- Prepare Modelfile Content
- Print System Info
- Push Model Gguf
- run
- Save Model Gguf
- Save Model Merged
- Train Model
- Vision Inference
- Create And Push Ollama Model
- Load Config
- Prepare Model
- Prepare Modelfile Content
- Push Model Gguf
- Save Model Merged
- upload
- generate
- Merge With Existing Workflow
- Recommend Pattern
- Recommend Pattern LLM
- agent
- Auth Callback
- Check Module Available
- Create Agent Scheduler
- Create Scheduler
- Disable Crewai Telemetry
- embedding
- Formatting Prompts Func
- Get All Available Tools
- Get Registered Agents
- Get Tools For Task
- main
- noop
- On Chat Resume
- On Modify
- On Run
- On Settings Update
- output
- profile
- Profile API
- Profile API Async
- Profile Async
- Profile Detailed
- Profile Imports
- Profile Lines
- Recommend Agent Count
- Run Agents
- Safe Format
- Sanitize Agent Name For Autogen V4
- Set Profiles
- Start Chat
- Start Chat Server
- task
- Time Import
- Tokenize Function
Functions
First Token • AI Agents Framework
first_token: Mark time to first token.
⌘I