ChatGPT finished 21/25, Claude 20/25 - and they won very different rounds.
ChatGPT obeyed the instructions. A 200-word essay came back at 185 words, "3 bullet points" came back as three short bullets, and the 3-line resume summary was ready to paste with no placeholders.
Claude won the two rounds where the reasoning was harder. Its prime-checking function skipped even numbers and stepped through odd divisors only, and its maths answer checked itself - it took 20% back off the result to prove it, then warned about the usual mistake of adding 20% of the sale price to get Rs 960.
The full scorecard:
| Prompt | ChatGPT Free | Claude Free | Winner |
|---|---|---|---|
| Essay (200 words) | 4/5 | 3/5 | ChatGPT |
| Coding (prime function) | 4/5 | 5/5 | Claude |
| Summary (3 bullets) | 5/5 | 4/5 | ChatGPT |
| Maths (20% discount) | 4/5 | 5/5 | Claude |
| Resume (3 lines) | 4/5 | 3/5 | ChatGPT |
| Total | 21/25 | 20/25 | ChatGPT |
A few things worth saying about the method, because it decides how much weight the numbers can carry:
- Five prompts, one new chat each, no follow-ups, no retries, nothing edited after seeing an answer.
- Both on the free plan, on 16 November 2026 between 1:05 and 2:20 AM IST. Claude showed Sonnet 5.5 on Medium; ChatGPT Free did not display a model name.
- Every answer was scored 1-5 on three things: was it correct, did it follow the instruction, and how much would I have to edit before using it.
- AI answers vary from run to run, so a second run could score differently. This is one run, on one night, with five prompts.
- I did not hit a usage limit on either tool. Five messages is nowhere near enough to find one.
The two failures are worth more than the totals. Claude ignored a word limit it was given, and returned a resume summary full of square-bracket placeholders. ChatGPT's coding answer was correct but padded, and its maths working did not verify itself.
Every screenshot is in the original post, unedited: https://voxahq.in/tests/chatgpt-free-vs-claude-free
Originally published at voxahq.in
Top comments (0)