DeepSeek V4.1 Flash 1017 scores 45 on AA Intelligence Index, ranks #2, runs at 225 tok/s
ArtificialAnlys · x · 2026-10-20
Artificial Analysis published full benchmark results for DeepSeek V4.1 Flash 1017 (Max): it scores 45 on the Intelligence Index v4.3.2, ranking #2 out of 121 models (median: 19).
Key data points:
- Speed: 225 output tokens/s, ranked 6/79 — first-tier
- Pricing: $0.30/M input, $1.20/M output tokens, 98% cache discount, $0.26 per Index task — moderately cheap
- Specs: 552B total / 16B active MoE, text+image input, 1M-token context, open weights
- Weakness: very verbose — 210M tokens generated during evals vs 140M median (verbosity rank 39/47)
Overall a top-tier open-weight model on intelligence and speed at reasonable pricing, though verbosity inflates real-world cost.
Related event: DeepSeek V4.1-Flash-1017: 'Minor Update' Tops Open-Weight Leaderboards(12 posts)→
More from Models
- lintool's Gaggle Matches SFT'd Qwen With Just a Handful of GPUs on nanochat Backbone — lintool · 2026-10-20
- Report: Liang Wenfeng planned independent DeepSeek compute chips before Huawei persuaded him otherwise — teortaxesTex · 2026-10-20
- Smallest AI's full voice stack hits 26.8% completion at $0.048/call on 300 hard calls — auto_grad_ · 2026-10-20
- NVIDIA's Dory1.0-4B Looped Model Lets You Pick Inference Depth at Runtime — lmoroney · 2026-10-20
- OpenAI dropped 722 AI-generated math papers in a day; can humans read them? — songyoupeng · 2026-10-20
- JevBench adds 34 decision models in four days; Nimble-v4-max price cut to $0.04/1M tokens — airesearch12 · 2026-10-20