Skip to content
← All frontiers

📚 Knowledge Q&A Assistant frontier

As of November 16, 2026, the highest Knowledge Q&A Assistant score on this frontier is Claude Fable 5.2 at 100.0, with an estimated monthly cost of $2,640.00.

An internal help desk or search assistant that answers questions from documents and general knowledge. Factual accuracy (SimpleQA) leads, with long-context retrieval for pasted documents, broad everyday ability, human preference for clear answers, and a slice of hard expert questions. Cost is weighted for steady daily volume.

The y axis is this scenario's weighted score; the x axis starts at estimated monthly cost for 2,000 requests a day. Switch axes to compare against speed, latency or context. See the full ranking.

On the frontier

Models on the Pareto frontier, highest score first
ModelScoreEst. $/month
Claude Fable 5.2Anthropic100.0$2,640
Muse WonderMeta89.9$1,320
Muse Spark 1.3Meta89.8$353
Gemini 4 CarbonGoogle87.0$330
Gemini 2.5 ProGoogle78.7$263
DeepSeek R1DeepSeek78.1$155
Qwen3 235BAlibaba (Qwen)77.1$49
DeepSeek V4.1 FlashDeepSeek63.6$38
Gemini 1.5 FlashGoogle62.0$23
GPT-6 LunaOpenAI46.2$17
DeepSeek V4 Flash 0731DeepSeek40.5$16
GPT-5 nanoOpenAI17.1$11