agent/agent-framework
1
0
Fork 0
mirror of https://github.com/microsoft/agent-framework.git synced 2026-06-16 21:04:09 +08:00
Code Issues Packages Projects Releases Wiki Activity

update link (#1051)

Browse Source
This commit is contained in:
Eric Zhu
2025-10-01 04:18:11 -07:00
committed by GitHub
Unverified
parent 5f1417ab94
commit 08b602c290
4 changed files with 33 additions and 25 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/packages/lab/README.md
+4 -2
View File
@@ -44,10 +44,12 @@ This structure maintains a single PyPI package `agent-framework-lab` while suppo
## Installation
Install the base lab package:
Install from source:
```bash
pip install agent-framework-lab
git clone https://github.com/microsoft/agent-framework.git
cd agent-framework/python/packages/lab
pip install -e .
```
For details on installing individual modules, see their respective README files listed above.
python/packages/lab/gaia/README.md
+4 -2
View File
@@ -7,10 +7,12 @@ It includes built-in benchmarks as well as utilities for running custom evaluati
## Setup
Install the agent-framework-lab package with GAIA dependencies:
Install from source with GAIA dependencies:
```bash
pip install "agent-framework-lab[gaia]"
git clone https://github.com/microsoft/agent-framework.git
cd agent-framework/python/packages/lab
pip install -e ".[gaia]"
```
Set up Hugging Face token:
python/packages/lab/lightning/README.md
+6 -4
View File
@@ -8,20 +8,22 @@ This package enables you to train and fine-tune agents using advanced RL algorit
## Installation
Install the agent-framework-lab package with Lightning dependencies:
Install from source with Lightning dependencies:
```bash
pip install "agent-framework-lab[lightning]"
git clone https://github.com/microsoft/agent-framework.git
cd agent-framework/python/packages/lab
pip install -e ".[lightning]"
```
### Optional Dependencies
```bash
# For math-related training
pip install agent-framework-lab[lightning,math]
pip install -e ".[lightning,math]"
# For tau2 benchmarking
pip install agent-framework-lab[lightning,tau2]
pip install -e ".[lightning,tau2]"
```
To prepare for RL training, you'll also need to install dependencies like PyTorch, Ray, and vLLM. See the [Agent-lightning setup instructions](https://github.com/microsoft/agent-lightning) for more details.
python/packages/lab/tau2/README.md
+19 -17
View File
@@ -13,20 +13,22 @@ Each evaluation runs a multi-turn conversation where the user simulator presents
## Supported Domains
| Domain | Status | Description |
|--------|--------|-------------|
| **airline** | ✅ Supported | Customer service for airline booking, changes, and support |
| **retail** | 🚧 In Development | E-commerce customer support scenarios |
| **telecom** | 🚧 In Development | Telecommunications service support |
| Domain | Status | Description |
| ----------- | ----------------- | ---------------------------------------------------------- |
| **airline** | ✅ Supported | Customer service for airline booking, changes, and support |
| **retail** | 🚧 In Development | E-commerce customer support scenarios |
| **telecom** | 🚧 In Development | Telecommunications service support |
*Note: Currently only the airline domain is fully supported.*
_Note: Currently only the airline domain is fully supported._
## Installation
Install the agent-framework-lab package with TAU2 dependencies:
Install from source with TAU2 dependencies:
```bash
pip install "agent-framework-lab[tau2]"
git clone https://github.com/microsoft/agent-framework.git
cd agent-framework/python/packages/lab
pip install -e ".[tau2]"
```
Download data from [Tau2-Bench](https://github.com/sierra-research/tau2-bench):
@@ -104,15 +106,15 @@ python samples/run_benchmark.py --max-steps 20
The following results are reproduced from our implementation of τ²-bench with `samples/run_benchmark.py`. It shows the average success rate over the dataset of 50 tasks.
| Agent Model | User Model | Success Rate |
|-------------|------------|----------|
| gpt-5 | gpt-4.1 | 62.0% |
| gpt-5-mini | gpt-4.1 | 52.0% |
| gpt-4.1 | gpt-4.1 | 60.0% |
| gpt-4.1-mini | gpt-4.1 | 50.0% |
| gpt-4.1 | gpt-4o-mini | 42.0% |
| gpt-4o | gpt-4.1 | 42.0% |
| gpt-4o-mini | gpt-4.1 | 26.0% |
| Agent Model | User Model | Success Rate |
| ------------ | ----------- | ------------ |
| gpt-5 | gpt-4.1 | 62.0% |
| gpt-5-mini | gpt-4.1 | 52.0% |
| gpt-4.1 | gpt-4.1 | 60.0% |
| gpt-4.1-mini | gpt-4.1 | 50.0% |
| gpt-4.1 | gpt-4o-mini | 42.0% |
| gpt-4o | gpt-4.1 | 42.0% |
| gpt-4o-mini | gpt-4.1 | 26.0% |
## Advanced Usage