[Obs AI Assistant] Add Llama 4 Maverick model ratings to the LLM performance matrix #3825
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
🔍 Preview links for changed docs |
viduni94
reviewed
Nov 5, 2025
| | OpenAI | **gpt-oss-20b** | Poor | Poor | Great | Poor | Good | Poor | Good | Good | | ||
| | OpenAI | **gpt-oss-120b** | Excellent | Poor | Great | Great | Excellent | Good | Good | Excellent | | ||
| | Meta | **Llama-3.3-70B-Instruct** | Excellent | Good | Great | Excellent | Excellent | Good | Good | Excellent | | ||
| | Meta | **Llama-4-Maverick-17B** | Good | Good | Good | Excellent | Excellent | Good | Good | Great | |
Contributor
There was a problem hiding this comment.
Let's stick to the HugginFace model name for consistency.
Suggested change
| | Meta | **Llama-4-Maverick-17B** | Good | Good | Good | Excellent | Excellent | Good | Good | Great | | |
| | Meta | **Llama-4-Maverick-17B-128E-Instruct** | Good | Good | Good | Excellent | Excellent | Good | Good | Great | |
viduni94
reviewed
Nov 5, 2025
| | OpenAI | **gpt-oss-20b** | Poor | Poor | Great | Poor | Good | Poor | Good | Good | | ||
| | OpenAI | **gpt-oss-120b** | Excellent | Poor | Great | Great | Excellent | Good | Good | Excellent | | ||
| | Meta | **Llama-3.3-70B-Instruct** | Excellent | Good | Great | Excellent | Excellent | Good | Good | Excellent | | ||
| | Meta | **Llama-4-Maverick-17B** | Good | Good | Good | Excellent | Excellent | Good | Good | Great | |
Contributor
There was a problem hiding this comment.
Most of the results are very different to llama 3.3 - Did you run into the token limit issue a lot ?
viduni94
reviewed
Nov 5, 2025
| | OpenAI | **gpt-oss-20b** | Poor | Poor | Great | Poor | Good | Poor | Good | Good | | ||
| | OpenAI | **gpt-oss-120b** | Excellent | Poor | Great | Great | Excellent | Good | Good | Excellent | | ||
| | Meta | **Llama-3.3-70B-Instruct** | Excellent | Good | Great | Excellent | Excellent | Good | Good | Excellent | | ||
| | Meta | **Llama-4-Maverick-17B** | Good | Good | Good | Excellent | Excellent | Good | Good | Great | |
Contributor
There was a problem hiding this comment.
The screenshot says Excellent for contextual insights, but noticed that it's specified as Good here
viduni94
approved these changes
Nov 6, 2025
Contributor
viduni94
left a comment
viduni94
left a comment
There was a problem hiding this comment.
LGTM 🎉
Thanks @yuliia-fryshko
viduni94
changed the title
mdbirnstiehl
approved these changes
Nov 6, 2025
pmoust
approved these changes
Nov 6, 2025
Member
pmoust
left a comment
pmoust
left a comment
There was a problem hiding this comment.
lgtm - thanks both for the work here and the reviews
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Closes https://github.com/elastic/obs-ai-assistant-team/issues/373
This PR adds the evaluation ratings for Llama 4 Maverick based on evaluation results.
I had to run the evaluation a couple of times, since the first run had some rate limit errors. It turns out that OpenRouter routes requests across multiple providers, each with different context limits - I added more details here.
In the latest run, I was lucky and didn’t get any errors.