Versions of the OpenAssist Dataset for chat fine-tuning different models. See https://youtu.be/71x8EMrB0Gc for a full video run through of methods.
AI & ML interests
Fine-tuning, Training and Deploying Large Language, Vision and Audio models.
Recent Activity
View all activity
Organization Card
last updated Feb 26th 2024
Trelis LTD is a research company based in Dublin, founded by Ronan McGovern
📚Watch Large Language, Vision and Audio Model Tutorials on YouTube
🛠 Explore Fine-tuning, Inference, Vision, Audio, and Evaluation Tools
Models fine-tuned for function-calling
-
Trelis/Meta-Llama-3-70B-Instruct-function-calling
Text Generation • 71B • Updated • 12 • 22 -
Trelis/Meta-Llama-3-8B-Instruct-function-calling
Text Generation • 8B • Updated • 1.07k • • 44 -
Trelis/openchat_3.5-function-calling-v3
Text Generation • 7B • Updated • 20 -
Trelis/Llama-2-7b-chat-hf-function-calling-v3
Text Generation • 7B • Updated • 204 • 41
Versions of the OpenAssist Dataset for chat fine-tuning different models. See https://youtu.be/71x8EMrB0Gc for a full video run through of methods.
Models fine-tuned for function-calling
-
Trelis/Meta-Llama-3-70B-Instruct-function-calling
Text Generation • 71B • Updated • 12 • 22 -
Trelis/Meta-Llama-3-8B-Instruct-function-calling
Text Generation • 8B • Updated • 1.07k • • 44 -
Trelis/openchat_3.5-function-calling-v3
Text Generation • 7B • Updated • 20 -
Trelis/Llama-2-7b-chat-hf-function-calling-v3
Text Generation • 7B • Updated • 204 • 41
spaces
5
Sleeping
Leaderboard Yourbench Trelis Yourbench-ait2
📈
Display benchmark results for a task
Sleeping
Leaderboard Yourbench Trelis Yourbench-ait
👀
Display and analyze leaderboard results
Sleeping
Leaderboard Yourbench Trelis Yourbench-touch-rugby-2
😻
Display auto-generated leaderboard and analyze task performance
Sleeping
Leaderboard Yourbench Trelis Touch-rugby-yourbench
🦀
models
356
Trelis/Qwen3-4B_ds-arc-agi-1-partial-100-c1542
Text Generation
•
4B
•
Updated
Trelis/Qwen3-4B_ds-arc-agi-1-partial-100-c771
Text Generation
•
4B
•
Updated
Trelis/Qwen3-4B_ds-arc-agi-1-partial-100-c8
Text Generation
•
4B
•
Updated
Trelis/Qwen3-4B_ds-arc-agi-1-partial-100-c4
Text Generation
•
4B
•
Updated
Trelis/Qwen3-4B_ds-arc-agi-1-partial-100_test-c8
Text Generation
•
4B
•
Updated
Trelis/Qwen3-4B_ds-arc-agi-1-partial-100_test-c4
Text Generation
•
4B
•
Updated
Trelis/Qwen3-4B_ds-arc-agi-1-perfect-50_test-c8
Text Generation
•
4B
•
Updated
•
2
Trelis/Qwen3-4B_ds-arc-agi-1-perfect-50_test-c4
Text Generation
•
4B
•
Updated
•
2
Trelis/Qwen3-4B_ds-arc-agi-2-perfect-50_test-c8
Text Generation
•
4B
•
Updated
•
2
Trelis/Qwen3-4B_ds-arc-agi-2-perfect-50_test-c4
Text Generation
•
4B
•
Updated
•
2
datasets
164
Trelis/arc-agi-1-partial-100
Viewer
•
Updated
•
24.7k
Trelis/arc-agi-1-perfect-50
Viewer
•
Updated
•
10.3k
•
6
Trelis/arc-agi-2-perfect-50
Viewer
•
Updated
•
15.5k
•
20
Trelis/arc-agi-2-partial-100
Viewer
•
Updated
•
50.5k
•
16
Trelis/arc-agi-1-blended-fake-ttt-50
Viewer
•
Updated
•
27.2k
•
13
Trelis/arc-agi-1-eval-programs-best-length-filtered-25-all
Viewer
•
Updated
•
9.98k
•
14
Trelis/arc-agi-1-eval-programs-best-length-filtered-25-partialplus
Viewer
•
Updated
•
4.09k
•
14
Trelis/arc-agi-1-train-programs-best-length-filtered-250
Viewer
•
Updated
•
59.2k
•
33
Trelis/arc-programs-50-full-200-incorrect
Viewer
•
Updated
•
92k
•
17
Trelis/arc-programs-50-full-200-partial
Viewer
•
Updated
•
69.5k
•
39