THE SCOREBOARD FOR WHAT'S NEXTAI + QUANTUM, SCORED. COMPARED. EXPLAINED.
AIQUANTUMSCORE.

OUR PROPRIETARY RESEARCH BENCHMARK

REVAL™
research evaluation.

Research Evaluation Index measures real-world AI research on a 0–100 scale. Scores are published only after actual evaluations and independent grading.

Methodology & weights →
ModelREVAL™
GPT-6 AstraNot tested
GPT-6 SolNot tested
GPT-6 LunaNot tested
GPT-6.1 SolNot tested
GPT-5.6 SolNot tested
GPT-5.6 TerraNot tested
GPT-5.6 LunaNot tested
Claude Fable 5.1Not tested
Claude Opus 5.5Not tested
Claude Sonnet 5.5Not tested
Claude Haiku 4.5Not tested
Gemini 3.8 FlashNot tested
Gemini 3.7 FlashNot tested
Gemini 3.6 FlashNot tested
Gemini 3.5 FlashNot tested
Gemini 3.5 Flash-LiteNot tested
Gemini 3.1 Pro PreviewNot tested
Grok 4.6Not tested
Grok 4.5Not tested
DeepSeek V4.1 FlashNot tested
DeepSeek V4 ProNot tested
Mistral Medium 3.5Not tested
Mistral Small 4Not tested
Mistral Large 3Not tested
Llama 4 ScoutNot tested
Llama 4 MaverickNot tested
Qwen3.5-397B-A17BNot tested
Qwen3.5-122B-A10BNot tested
gpt-4.1Not tested
gpt-4.1-miniNot tested
gpt-4.1-nanoNot tested
gpt-5Not tested
gpt-5-miniNot tested
gpt-5-nanoNot tested
gpt-5.1Not tested
gpt-5.2Not tested
gpt-5.4-miniNot tested
gpt-5.4-nanoNot tested
Gemini 4 ArgonNot tested
Muse WonderNot tested
Muse Wonder FullNot tested