Comments (2)
Thanks for the report. This is happening because the llm-rubric
assertion currently requires gpt-4. I'll aim to get a fix in tonight or tomorrow
from promptfoo.
Your best bet is to use gpt-3.5-turbo and override the rubric grader. As of 0.19.3, you can do this in one of several ways:
- Using the
--grader
CLI option:promptfoo eval --grader openai:gpt-3.5-turbo
- Using
test.options
ordefaultTest.options
on a per-test or testsuite basis:defaultTest: options: provider: gpt-3.5-turbo tests: - description: Use LLM to evaluate output assert: - type: llm-rubric value: Is spoken like a pirate
- Using
assertion.provider
on a per-assertion basis:tests: - description: Use LLM to evaluate output assert: - type: llm-rubric value: Is spoken like a pirate provider: gpt-3.5-turbo
from promptfoo.
Related Issues (20)
- scenarios tied to prompts and not to vars HOT 12
- prompts from file different behavior than prompts from config HOT 2
- assert type similar should consider custom provider js file HOT 3
- Eval fails when a prompt string ends with a period "." HOT 3
- Error running example classifier assertion HOT 3
- Group tests into categories HOT 1
- Is there a way to Separate Tests by Prompt?
- Evalulate Multiple Config Files Separately Instead of Combining them HOT 2
- ReDoS Vulnerability in Async Dependency in promptfoo Library HOT 2
- Request: web UI showing `promptfoo`'s version
- Progress bar not matching number of `tests`
- Is there a way to assert only a part of the LLMresponse ? HOT 2
- Issue while using message arrays as a prompts with HTTP provider
- Slow evaluation with inner provider loop HOT 4
- Providing different prompts per model always takes arguments of last provider HOT 1
- bug: is-sql assertion allowedColumns does not work HOT 1
- Check CLI arguments like --output before executing evals HOT 2
- Unable to share eval HOT 1
- Request: link in variables HOT 3
- Request: streaming into `eval --output` HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from promptfoo.