You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+11-9Lines changed: 11 additions & 9 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -6,15 +6,17 @@ Crab is a framework for building LLM agent benchmark environments in a Python-ce
6
6
7
7
#### Key Features
8
8
9
-
* 🌐 Cross-platform
10
-
* Build agent environments in memory, hosted through a docker environment, installed a virtual machine, or as a distributed physical machine, as long as they can be accessed by Python functions.
11
-
* Let the agent access all the environments in the same time through a unified interface.
12
-
* ⚙ ️Easy-to-use Configuration
13
-
* Add a new action by simply adding a `@action` decorator on a Python function.
14
-
* Deine the environment by integrating several actions together.
15
-
* 📐 Novel Benchmarking Suite
16
-
* Define tasks and the corresponding evlauators in an intuitive Python-native way.
17
-
* Introduce a novel graph evaluator method proving fine-grained metrics.
9
+
🌐 Cross-platform
10
+
* Create build agent environments that support various deployment options including in-memory, Docker-hosted, virtual machines, or distributed physical machines, provided they are accessible via Python functions.
11
+
* Let the agent access all the environments in the same time through a unified interface.
12
+
13
+
⚙ ️Easy-to-use Configuration
14
+
* Add a new action by simply adding a `@action` decorator on a Python function.
15
+
* Deine the environment by integrating several actions together.
16
+
17
+
📐 Novel Benchmarking Suite
18
+
* Define tasks and the corresponding evlauators in an intuitive Python-native way.
19
+
* Introduce a novel graph evaluator method providing fine-grained metrics.
0 commit comments