sonar-reasoning LMArena Search Arena 24K
deepseek v3 0324 gpqa benchmark? Answer in extremely short.
- User
deepseek v3 0324 gpqa benchmark? Answer in extremely short.
- Assistant
DeepSeek-V3 0324 achieved a GPQA score of 68.4, improving by +9.3 over previous versions[5][9].
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.