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

feat(py): Added collecting of usage info for Ollama API #2361

Merged
merged 2 commits into from
Mar 20, 2025
Merged

feat(py): Added collecting of usage info for Ollama API #2361

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e4d8a73
feat(py): Added collecting of usage info for Ollama API
kirgrim Mar 17, 2025
c6ee651
Merge branch 'main' into hrymailo/feat/ollama-usage-stats-in-response
kirgrim Mar 20, 2025
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
46 changes: 40 additions & 6 deletions py/plugins/ollama/src/genkit/plugins/ollama/models.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,15 +4,17 @@
import mimetypes
from typing import Literal

import ollama as ollama_api
from pydantic import BaseModel, Field, HttpUrl

import ollama as ollama_api
from genkit.ai.model import get_basic_usage_stats
from genkit.core.action import ActionRunContext
from genkit.core.typing import (
GenerateRequest,
GenerateResponse,
GenerateResponseChunk,
GenerationCommonConfig,
GenerationUsage,
Media,
MediaPart,
Message,
Expand Down Expand Up @@ -61,12 +63,12 @@ async def generate(
content = [TextPart(text='Failed to get response from Ollama API')]

if self.model_definition.api_type == OllamaAPITypes.CHAT:
chat_response = await self._chat_with_ollama(
api_response = await self._chat_with_ollama(
request=request, ctx=ctx
)
if chat_response:
if api_response:
content = self._build_multimodal_chat_response(
chat_response=chat_response,
chat_response=api_response,
)
elif self.model_definition.api_type == OllamaAPITypes.GENERATE:
api_response = await self._generate_ollama_response(
Expand All @@ -75,16 +77,32 @@ async def generate(
if api_response:
content = [TextPart(text=api_response.response)]
else:
LOG.error(f'Unresolved API type: {self.model_definition.api_type}')
raise ValueError(
f'Unresolved API type: {self.model_definition.api_type}'
)

if self.is_streaming_request(ctx=ctx):
content = []

response_message = Message(
role=Role.MODEL,
content=content,
)

basic_generation_usage = get_basic_usage_stats(
input_=request.messages,
response=response_message,
)

return GenerateResponse(
message=Message(
role=Role.MODEL,
content=content,
)
),
usage=self.get_usage_info(
basic_generation_usage=basic_generation_usage,
api_response=api_response,
),
)

async def _chat_with_ollama(
Expand Down Expand Up @@ -277,3 +295,19 @@ def _to_ollama_role(
@staticmethod
def is_streaming_request(ctx: ActionRunContext | None) -> bool:
return ctx and ctx.is_streaming

@staticmethod
def get_usage_info(
basic_generation_usage: GenerationUsage,
api_response: ollama_api.GenerateResponse | ollama_api.ChatResponse,
) -> GenerationUsage:
if api_response:
basic_generation_usage.input_tokens = (
api_response.prompt_eval_count or 0
)
basic_generation_usage.output_tokens = api_response.eval_count or 0
basic_generation_usage.total_tokens = (
basic_generation_usage.input_tokens
+ basic_generation_usage.output_tokens
)
return basic_generation_usage
Loading