What will Claude 3.5 Opus's reported 0-shot performance on GPQA Diamond be upon release?
Basic
1
Ṁ1002025
1D
1W
1M
ALL
14%
[0%, 60%)
32%
[60%, 70%)
32%
[70%, 80%)
14%
[80%, 90%)
8%
[90%, 100%]
Rather than yet another market speculating on the exact date Claude 3.5 Opus's, I find it more interesting to see how good people's projections of capabilities are at this point. Towards this, I'm curious to get estimate of Claude 3.5 Opus's GPQA Diamond performance.
For context, Claude 3.5 Sonnet achieved 59.4% accuracy, and o1 (unreleased version reportedly achieves 77.3% accuracy (pass@1, rightmost).
If GPQA Diamond performance for Claude 3.5 Opus isn't reported within 3 months following release, either by Anthropic or a source I consider credible (e.g. the benchmark creators, Scale's benchmarking team, etc.), I'll resolve this N/A.
Note: I won't bet on this market.
This question is managed and resolved by Manifold.
Get
1,000
and3.00
Related questions
Related questions
Will we get a video of claude 3.5 Sonnet running a very single minded competent minecraft agent before December 2024?
71% chance
Will Claude 3.5 Opus be released in October?
4% chance
When will Claude 3.5 Opus be released?
Will Claude 3.5 Opus be available via API by end of 2025?
46% chance
Will Claude 3.5 Haiku be better than Claude 3 Opus?
29% chance
What will be the *first* ELO Rating of Claude 3.5 Opus in the LMSYS Arena?
Will Claude 3.5 Opus beat OpenAI's best released model on the arena.lmsys.org leaderboard?
23% chance
What ELO will new Claude Sonnet 3.5 get in LMArena?
When will Claude 4 be released?
Will Claude 3.5 Opus be able to draw me in tic-tac-toe while playing as O at least 1/3 of the time?
68% chance