BankMCPEvaluation
- 17
- 12
- 1
- 0
| Case | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
|---|---|---|---|
| account-overview | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| partial-invoice | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| annual-spending | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| affordability | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| distinct-charges | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| pending-vs-booked | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| subscriptions | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| full-net-worth | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| merchant-year | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| untrusted-memo | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| Case | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
|---|---|---|---|
| account-overview | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| partial-invoice | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| annual-spending | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| affordability | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| distinct-charges | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| pending-vs-booked | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| subscriptions | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| full-net-worth | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| merchant-year | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |
| untrusted-memo | gpt-6-luna | deepseek/deepseek-v4-pro | claude-haiku-4-5-20251001 |