@@ -72,16 +72,18 @@ Follow this link to view and download [example datasets](https://github.com/Azur
72
72
## Model Comparison Table
73
73
This table provides an overview of the models available
74
74
75
- | Model | Modalities | Techniques | Strengths |
76
- | ----------------------| ---------------| --------------| --------------------------------------|
75
+ | Model | Modalities | Techniques | Strengths |
76
+ | ----------------------| ---------------| --------------| -------------------------------------------------------------------- |
77
77
| GPT 4.1 | Text, Vision | SFT, DPO | Superior performance on sophisticated tasks, nuanced understanding |
78
- | GPT 4.1-mini | Text | SFT, DPO | Fast iteration, cost-effective, good for simple tasks |
79
- | GPT 4.1-nano | Text | SFT, DPO | Fast, cost-effective, and minimal resource usage |
80
- | o4-mini | Text | RFT | Reasoning model suited for complex logical tasks |
81
- | Phi 4 | Text | SFT | Cost effective option for simpler tasks |
82
- | Ministral 3B | Text | SFT | Low-cost option for faster iteration |
83
- | Mistral Nemo | Text | SFT | Balance between size and capability |
84
- | Mistral Large (2411) | Text | SFT | Most capable Mistral model, better for complex tasks |
78
+ | GPT 4.1-mini | Text | SFT, DPO | Fast iteration, cost-effective, good for simple tasks |
79
+ | GPT 4.1-nano | Text | SFT, DPO | Fast, cost-effective, and minimal resource usage |
80
+ | GPT 4o | Text, Vision | SFT, DPO | Previous generation flagship model for complex tasks |
81
+ | GPT 4o-mini | Text | SFT | Previous generation small model for simple tasks |
82
+ | o4-mini | Text | RFT | Reasoning model suited for complex logical tasks |
83
+ | Phi 4 | Text | SFT | Cost effective option for simpler tasks |
84
+ | Ministral 3B | Text | SFT | Low-cost option for faster iteration |
85
+ | Mistral Nemo | Text | SFT | Balance between size and capability |
86
+ | Mistral Large (2411) | Text | SFT | Most capable Mistral model, better for complex tasks |
85
87
86
88
## Get Started with Fine Tuning
87
89
0 commit comments