@@ -5,7 +5,7 @@ Two AI models take turns telling a knock-knock joke while measuring response tim
55## Models Used
66
77- ** Fast model** : ` gpt-4.1 ` - Delivers the joke setup and punchline
8- - ** Smart model** : ` claude-sonnet -4.5 ` - Responds to the joke
8+ - ** Smart model** : ` claude-opus -4.5 ` - Responds to the joke (premium model)
99
1010## Run
1111
@@ -21,28 +21,28 @@ npx tsx examples/knock-knock.ts
2121🚪 Knock-Knock Joke: Two Models Battle!
2222
2323Fast model: gpt-4.1
24- Smart model: claude-sonnet -4.5
24+ Smart model: claude-opus -4.5
2525
2626──────────────────────────────────────────────────
2727🏃 gpt-4.1: Knock knock!
28- 🧠 claude-sonnet -4.5: Who's there?
28+ 🧠 claude-opus -4.5: Who's there?
2929🏃 gpt-4.1: Olive
30- 🧠 claude-sonnet -4.5: Olive who?
30+ 🧠 claude-opus -4.5: Olive who?
3131🏃 gpt-4.1: Olive you and I miss you!
3232──────────────────────────────────────────────────
3333
3434⏱️ Response Times:
3535
36- 🏃 gpt-4.1 | Knock knock | 4542ms
37- 🧠 claude-sonnet -4.5 | Who's there? | 3533ms
38- 🏃 gpt-4.1 | Setup | 1428ms
39- 🧠 claude-sonnet -4.5 | Who? | 4321ms
40- 🏃 gpt-4.1 | Punchline | 2473ms
36+ 🏃 gpt-4.1 | Knock knock | 4490ms
37+ 🧠 claude-opus -4.5 | Who's there? | 2478ms
38+ 🏃 gpt-4.1 | Setup | 1289ms
39+ 🧠 claude-opus -4.5 | Who? | 2713ms
40+ 🏃 gpt-4.1 | Punchline | 1257ms
4141──────────────────────────────────────────────────
42- 🏃 gpt-4.1 total: 8443ms (3 turns)
43- 🧠 claude-sonnet -4.5 total: 7854ms (2 turns)
42+ 🏃 gpt-4.1 total: 7035ms (3 turns)
43+ 🧠 claude-opus -4.5 total: 5191ms (2 turns)
4444
45- 🏆 claude-sonnet -4.5 was faster!
45+ 🏆 claude-opus -4.5 was faster!
4646```
4747
4848## Customize
@@ -51,5 +51,5 @@ Edit the constants at the top of `knock-knock.ts` to try different models:
5151
5252``` typescript
5353const FAST_MODEL = " gpt-4.1" ;
54- const SMART_MODEL = " claude-sonnet -4.5" ;
54+ const SMART_MODEL = " claude-opus -4.5" ;
5555```
0 commit comments