Recap · 2025-01
AI leaderboard changes January 2025
As of Jan 28, 2025, 11 confirmed crown changes are recorded for January 2025.
Crowns
Confirmed crown changes
- Jan 28, 2025. General LLM: O1 and DeepSeek R1 are co-SOTA. Crown change confirmed by two consecutive snapshots or two independent primary sources. O1DeepSeek R1 · General LLM
- Jan 28, 2025. Reasoning: DeepSeek R1 takes the crown. Crown change confirmed by two consecutive snapshots or two independent primary sources. DeepSeek R1 · Reasoning
- Jan 28, 2025. General LLM: DeepSeek R1 takes the crown. Crown change confirmed by two consecutive snapshots or two independent primary sources. DeepSeek R1 · General LLM
- Jan 15, 2025. Reasoning: O1 and O1 Preview are co-SOTA. Crown change confirmed by two consecutive snapshots or two independent primary sources. O1O1 Preview · Reasoning
- Jan 15, 2025. Long context: O1 and Claude 3.5 Sonnet and O1 Preview are co-SOTA. Crown change confirmed by two consecutive snapshots or two independent primary sources. O1Claude 3.5 SonnetO1 Preview · Long context
- Jan 5, 2025. Image generation: Flux 1 Dev Fp8 takes the crown. First holder recorded for this category and slice. Flux 1 Dev Fp8 · Image generation
- Jan 5, 2025. General LLM: O1 takes the crown. Crown change confirmed by two consecutive snapshots or two independent primary sources. O1 · General LLM
- Jan 5, 2025. Reasoning: DeepSeek V3 and Llama 3.1 405B Instruct Bf16 and Llama 3.1 405B Instruct Fp8 are co-SOTA. Crown change confirmed by two consecutive snapshots or two independent primary sources. DeepSeek V3Llama 3.1 405B Instruct Bf16Llama 3.1 405B Instruct Fp8 · Reasoning
- Jan 5, 2025. Long context: DeepSeek V3 takes the crown. First holder recorded for this category and slice. DeepSeek V3 · Long context
- Jan 5, 2025. Image generation: Flux 1 Dev Fp8 takes the crown. First holder recorded for this category and slice. Flux 1 Dev Fp8 · Image generation
- Jan 5, 2025. General LLM: DeepSeek V3 takes the crown. Crown change confirmed by two consecutive snapshots or two independent primary sources. DeepSeek V3 · General LLM
Top 3
Models that entered the top 3
- Claude 3.5 Sonnet at #1 in Long context
- DeepSeek R1 at #1 in General LLM
- Flux 1 Dev Fp8 at #1 in Image generation
- O1 at #1 in General LLM
- O1 at #1 in Reasoning
- O1 at #1 in Long context
- O1 Preview at #1 in Reasoning
- O1 Preview at #1 in Long context
Benchmarks
Benchmarks added or saturated
Benchmark additions for this month are still empty.