Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[BFCL] Add gemini-1.5-pro-002, gemini-1.5-pro-002-FC, gemini-1.5-pro-001, gemini-1.5-pro-001-FC, gemini-1.5-flash-002, gemini-1.5-flash-002-FC, gemini-1.0-pro-002, gemini-1.0-pro-002-FC #658

Merged
merged 21 commits into from
Oct 5, 2024
Merged

[BFCL] Add gemini-1.5-pro-002, gemini-1.5-pro-002-FC, gemini-1.5-pro-001, gemini-1.5-pro-001-FC, gemini-1.5-flash-002, gemini-1.5-flash-002-FC, gemini-1.0-pro-002, gemini-1.0-pro-002-FC #658

Show file tree
Hide file tree
Changes from 18 commits
Commits
Show all changes
21 commits
Select commit Hold shift + click to select a range
6703628
update gemini handler to support prompting mode
HuanzhiMao Sep 24, 2024
7af2763
update decoding method
HuanzhiMao Sep 24, 2024
956fe18
update model metadata
HuanzhiMao Sep 24, 2024
f0f717e
fix tool execution result format
HuanzhiMao Sep 24, 2024
ca99db6
better handle errors when gemini return no response
HuanzhiMao Sep 24, 2024
7b4de3e
Merge branch 'main' into gemini
HuanzhiMao Sep 25, 2024
5b13bbe
add gemini-1.5-pro-002
HuanzhiMao Sep 26, 2024
e3d5e1f
patch some Vertex AI library bugs that affect decoding for FC models …
HuanzhiMao Sep 26, 2024
558ea03
Update gemini.py to handle Vertex AI library bug
HuanzhiMao Sep 29, 2024
ed11381
clean up
HuanzhiMao Sep 29, 2024
abdd367
better handle enum field when param type is not string
HuanzhiMao Sep 29, 2024
29f79d4
No `optional` field in parameter schema
HuanzhiMao Sep 29, 2024
2b6359e
handle `minItems`, `maxItems`
HuanzhiMao Sep 29, 2024
fd5e483
update ast decoding for gemini FC version
HuanzhiMao Sep 29, 2024
4eee739
better handler prompt mode output parsing; remove trailing \n
HuanzhiMao Sep 29, 2024
a67150c
Merge branch 'main' into gemini
HuanzhiMao Sep 29, 2024
4553476
remove leading and trailing \n
HuanzhiMao Sep 29, 2024
f2dc473
use Gemini GA version models
HuanzhiMao Oct 3, 2024
6af5929
update cost; add flash-001
HuanzhiMao Oct 4, 2024
d44d13f
Merge branch 'main' into gemini
HuanzhiMao Oct 4, 2024
99e2942
Merge branch 'main' into gemini
HuanzhiMao Oct 5, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 9 additions & 0 deletions berkeley-function-call-leaderboard/CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,15 @@ All notable changes to the Berkeley Function Calling Leaderboard will be documen
- `microsoft/Phi-3-mini-128k-instruct`
- `microsoft/Phi-3-mini-4k-instruct`
- [Sept 25, 2024] [#660](https://github.com/ShishirPatil/gorilla/pull/660): Bug fix in `parse_nested_value` function to handle nested dictionary values properly.
- [Sept 24, 2024] [#648](https://github.com/ShishirPatil/gorilla/pull/648): Add the following new models to the leaderboard:
- `gemini-1.5-pro-002`
- `gemini-1.5-pro-002-FC`
- `gemini-1.5-pro-001`
- `gemini-1.5-pro-001-FC`
- `gemini-1.5-flash-002`
HuanzhiMao marked this conversation as resolved.
Show resolved Hide resolved
- `gemini-1.5-flash-002-FC`
- `gemini-1.0-pro-002`
- `gemini-1.0-pro-002-FC`
- [Sept 19, 2024] [#644](https://github.com/ShishirPatil/gorilla/pull/644): BFCL V3 release:
- Introduce new multi-turn dataset and state-based evaluation metric
- Separate ast_checker and executable_checker for readability
Expand Down
9 changes: 6 additions & 3 deletions berkeley-function-call-leaderboard/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,9 +132,12 @@ Below is _a table of models we support_ to run our leaderboard evaluation agains
|databrick-dbrx-instruct | Prompt|
|deepseek-ai/deepseek-coder-6.7b-instruct 💻| Prompt|
|firefunction-{v1,v2}-FC | Function Calling|
|gemini-1.0-pro-FC | Function Calling|
|gemini-1.5-pro-preview-{0409,0514}-FC | Function Calling|
|gemini-1.5-flash-preview-0514-FC | Function Calling|
|gemini-1.0-pro-{001,002}-FC | Function Calling|
|gemini-1.0-pro-{001,002} | Prompt|
|gemini-1.5-pro-{001,002}-FC | Function Calling|
|gemini-1.5-pro-{001,002} | Prompt|
|gemini-1.5-flash-{001,002}-FC | Function Calling|
|gemini-1.5-flash-{001,002} | Prompt|
|glaiveai/glaive-function-calling-v1 💻| Function Calling|
|gpt-3.5-turbo-0125-FC| Function Calling|
|gpt-3.5-turbo-0125| Prompt|
Expand Down
44 changes: 34 additions & 10 deletions berkeley-function-call-leaderboard/bfcl/eval_checker/model_metadata.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -263,27 +263,51 @@
"Fireworks",
"Apache 2.0",
],
"gemini-1.5-pro-preview-0514-FC": [
"Gemini-1.5-Pro-Preview-0514 (FC)",
"gemini-1.5-pro-002": [
"Gemini-1.5-Pro-002 (Prompt)",
"https://deepmind.google/technologies/gemini/pro/",
"Google",
"Proprietary",
],
"gemini-1.5-flash-preview-0514-FC": [
"Gemini-1.5-Flash-Preview-0514 (FC)",
"gemini-1.5-pro-002-FC": [
"Gemini-1.5-Pro-002 (FC)",
"https://deepmind.google/technologies/gemini/pro/",
"Google",
"Proprietary",
],
"gemini-1.5-pro-001": [
"Gemini-1.5-Pro-001 (Prompt)",
"https://deepmind.google/technologies/gemini/pro/",
"Google",
"Proprietary",
],
"gemini-1.5-pro-001-FC": [
"Gemini-1.5-Pro-001 (FC)",
"https://deepmind.google/technologies/gemini/pro/",
"Google",
"Proprietary",
],
"gemini-1.5-flash-002": [
"Gemini-1.5-Flash-002 (Prompt)",
"https://deepmind.google/technologies/gemini/flash/",
"Google",
"Proprietary",
],
"gemini-1.5-pro-preview-0409-FC": [
"Gemini-1.5-Pro-Preview-0409 (FC)",
"https://deepmind.google/technologies/gemini/#introduction",
"gemini-1.5-flash-002-FC": [
"Gemini-1.5-Flash-002 (FC)",
"https://deepmind.google/technologies/gemini/flash/",
"Google",
"Proprietary",
],
"gemini-1.0-pro-FC": [
"Gemini-1.0-Pro-001 (FC)",
"https://deepmind.google/technologies/gemini/#introduction",
"gemini-1.0-pro-002": [
"Gemini-1.0-Pro-002 (Prompt)",
"https://deepmind.google/technologies/gemini/pro/",
"Google",
"Proprietary",
],
"gemini-1.0-pro-002-FC": [
"Gemini-1.0-Pro-002 (FC)",
"https://deepmind.google/technologies/gemini/pro/",
"Google",
"Proprietary",
],
Expand Down
8 changes: 4 additions & 4 deletions berkeley-function-call-leaderboard/bfcl/model_handler/constant.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -128,10 +128,10 @@
"mistral-large-2407-FC",
"mistral-small-2402-FC",
"mistral-small-2402-FC",
"gemini-1.0-pro-FC",
"gemini-1.5-pro-preview-0409-FC",
"gemini-1.5-pro-preview-0514-FC",
"gemini-1.5-flash-preview-0514-FC",
"gemini-1.5-pro-002-FC",
"gemini-1.5-pro-001-FC",
"gemini-1.5-flash-002-FC",
"gemini-1.0-pro-002-FC",
"meetkai/functionary-small-v3.1-FC",
"meetkai/functionary-small-v3.2-FC",
"meetkai/functionary-medium-v3.1-FC",
Expand Down
16 changes: 12 additions & 4 deletions berkeley-function-call-leaderboard/bfcl/model_handler/handler_map.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,9 +56,14 @@
"firefunction-v1-FC": FireworksHandler,
"firefunction-v2-FC": FireworksHandler,
"Nexusflow-Raven-v2": NexusHandler,
"gemini-1.0-pro-FC": GeminiHandler,
"gemini-1.5-pro-preview-0514-FC": GeminiHandler,
"gemini-1.5-flash-preview-0514-FC": GeminiHandler,
"gemini-1.5-pro-002": GeminiHandler,
"gemini-1.5-pro-002-FC": GeminiHandler,
"gemini-1.5-pro-001": GeminiHandler,
"gemini-1.5-pro-001-FC": GeminiHandler,
"gemini-1.5-flash-002": GeminiHandler,
"gemini-1.5-flash-002-FC": GeminiHandler,
"gemini-1.0-pro-002": GeminiHandler,
"gemini-1.0-pro-002-FC": GeminiHandler,
"meetkai/functionary-small-v3.2-FC": FunctionaryHandler,
"meetkai/functionary-medium-v3.1-FC": FunctionaryHandler,
"databricks-dbrx-instruct": DatabricksHandler,
Expand Down Expand Up @@ -105,7 +110,10 @@
# "gpt-4-0613": OpenAIHandler,
# "claude-2.1": ClaudeHandler,
# "claude-instant-1.2": ClaudeHandler,
# "gemini-1.5-pro-preview-0409-FC": GeminiHandler,
# "gemini-1.0-pro-001": GeminiHandler,
# "gemini-1.0-pro-001-FC": GeminiHandler,
# "gemini-1.5-flash-001": GeminiHandler,
# "gemini-1.5-flash-001-FC": GeminiHandler,
# "meetkai/functionary-small-v3.1-FC": FunctionaryHandler,
# "mistral-tiny-2312": MistralHandler,
# "glaiveai/glaive-function-calling-v1": GlaiveHandler,
Expand Down
Loading