github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-22 13:42:25 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #12746] Benchmarking Local Ollama Models (Performance & Quality) - created. #8452

New Issue
Open
opened 2026-04-12 21:08:08 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-12 21:08:08 -05:00
Owner
Copy Link

Originally created by @cumhur on GitHub (Oct 22, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12746

Hello Ollama Team and Community,

First, I want to say thank you for the incredible work on Ollama. It has completely changed how many of us experiment with local LLMs.

To help myself and others better evaluate and compare the performance of different models, I've developed a Python-based benchmarking tool. I wanted to share it with the community in case it's useful, or perhaps to inspire future official features.

GitHub Repository: https://github.com/cumhur/Ollama_Local_LLM_Benchmark

This tool allows users to test multiple local models against a list of prompts. It has two key outputs:

  1. Performance Dashboard (.html): It generates a simple HTML dashboard (using Matplotlib) that visually compares key metrics like total response time, load duration, and tokens/second. This makes it easy to see performance differences at a glance.
  2. Detailed Report (.csv): It saves all benchmark data and the full text response from each model into a CSV file. This is crucial because it allows for secondary "quality" analysis (e.g., one could feed these responses to a more powerful model like GPT-4 or Claude 3 to score their accuracy or coherence).

I created this because I needed a consistent way to see how different models and configurations perform on my own hardware.

I believe a robust, integrated benchmarking feature would be a powerful addition to the Ollama ecosystem, helping users quickly find the best model (for speed and quality) for their specific needs.

Thank you for your time and for building such a great platform!

Originally created by @cumhur on GitHub (Oct 22, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12746 Hello Ollama Team and Community, First, I want to say thank you for the incredible work on Ollama. It has completely changed how many of us experiment with local LLMs. To help myself and others better evaluate and compare the performance of different models, I've developed a Python-based benchmarking tool. I wanted to share it with the community in case it's useful, or perhaps to inspire future official features. **GitHub Repository:** https://github.com/cumhur/Ollama_Local_LLM_Benchmark This tool allows users to test multiple local models against a list of prompts. It has two key outputs: 1. **Performance Dashboard (.html):** It generates a simple HTML dashboard (using Matplotlib) that visually compares key metrics like total response time, load duration, and tokens/second. This makes it easy to see performance differences at a glance. 2. **Detailed Report (.csv):** It saves all benchmark data *and the full text response* from each model into a CSV file. This is crucial because it allows for secondary "quality" analysis (e.g., one could feed these responses to a more powerful model like GPT-4 or Claude 3 to score their accuracy or coherence). I created this because I needed a consistent way to see how different models and configurations perform on my own hardware. I believe a robust, integrated benchmarking feature would be a powerful addition to the Ollama ecosystem, helping users quickly find the best model (for speed *and* quality) for their specific needs. Thank you for your time and for building such a great platform!
GiteaMirror added the feature request label 2026-04-12 21:08:08 -05:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label feature request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#8452
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?