[Feature]: Use openai 4o-mini as default instead of 3.5-turbo - cheaper and better

[jakob1379]

trafficstars

Description

Currently, OpenCommit uses GPT-3.5 as the default model for its functionality. However, this model is both more expensive and less capable than the newer GPT-4o-mini. GPT-4o-mini offers better performance for many tasks while significantly reducing costs.

---

Suggested Solution

Update the default model in OpenCommit to gpt-4o-mini.

Key benefits of this change:

• Cost Efficiency: GPT-4o-mini costs $0.15 per 1M input tokens and $0.60 per 1M output tokens, compared to GPT-3.5 Turbo's $1.50 per 1M input tokens and $2.00 per 1M output tokens.
• Improved Capabilities: GPT-4o-mini is optimized for smaller-scale applications, offering advanced features such as vision capabilities and better reasoning performance compared to GPT-3.5 Turbo.
• Accessibility: Reduces the financial barrier for contributors and users of OpenCommit.

---

Additional Context

• Model Pricing Comparison:

  • GPT-3.5 Turbo: $1.50 per 1M input tokens, $2.00 per 1M output tokens.
  • GPT-4o-mini: $0.15 per 1M input tokens, $0.60 per 1M output tokens.
  • Relative Savings: GPT-4o-mini is approximately 90% cheaper for inputs and 70% cheaper for outputs.

• Documentation: [OpenAI Pricing Page](https://openai.com/pricing)

• This change aligns with the project's goals of providing an efficient, cost-effective tool for contributors while leveraging state-of-the-art AI models.

---

References

• [OpenAI Pricing Information](https://openai.com/pricing)
• [GPT-4o-mini Model Capabilities](https://openai.com/gpt-4o-mini)