# DeepSeek V4 Flash 0731 在 ARC-AGI 基准上公布验证得分 - 来源：Hacker News 热门（buzzing.cc 中文翻译） - 作者：tosh - 发布时间：2026-08-08 03:10 - AIHOT 分数：49 - AIHOT 链接：https://aihot.virxact.com/items/cmsjcplur062vroo5x91oo2fw - 原文链接：https://arcprize.org/results/deepseek-v4-flash-0731 ## AI 摘要 DeepSeek V4 Flash 0731 在最大推理强度下，ARC-AGI-1 Semi-Private 得分 89.0%，每任务成本 $0.02；ARC-AGI-2 Semi-Private 得分 61.4%，每任务成本 $0.04。该模型在 ARC-AGI-1 和 ARC-AGI-2 上均提供 Max、High、Low 三档推理强度下的验证分数。 ## 正文 At max effort, DeepSeek V4 Flash 0731 scores 89.0% on ARC-AGI-1 Semi-Private at $0.02 per task and 61.4% on ARC-AGI-2 Semi-Private at $0.04 per task. ARC-AGI 2 leaderboard DeepSeek V4 Flash 0731 Verified scores VariantARC-AGI-1ARC-AGI-2ARC-AGI-3 Max 89.0% 61.4% — High 87.0% 56.0% — Low 84.0% 46.0% — Tasks & environments Pass/fail per reasoning level across each benchmark. ARC-AGI-2 Public Eval 120 tasks Task Max High Low 0934a4d8 ✗✓✗ 135a2760 ✓✓✓ 136b0064 ✓✓✓ 13e47133 ✗✗✗ 142ca369 ✗✓✗ 16b78196 ✗✗✗ 16de56c4 ✓✓✓ 1818057f ✓✓✓ 195c6913 ✓✗✗ 1ae2feb7 ✓✓✓ 20270e3b ✓✓✗ 20a9e565 ✗✗✗ 21897d95 ✗✗✗ 221dfab4 ✓✓✗ 247ef758 ✓✓✓ 269e22fb ✗✗✗ 271d71e2 ✗✓✗ 28a6681f ✓✗✗ 291dc1e1 ✓✓✗ 2b83f449 ✗✗✗ 2ba387bc ✓✓✓ 2c181942 ✓✓✗ 2d0172a1 ✗✗✗ 31f7f899 ✓✓✓ 332f06d7 ✓✗✗ 35ab12c3 ✓✓✓ 36a08778 ✓✓✓ 38007db0 ✓✓✓ 3a25b0d8 ✗✗✗ 3dc255db ✗✗✗ 3e6067c3 ✓✗✗ 409aa875 ✓✓✗ 446ef5d2 ✗✗✗ 45a5af55 ✓✓✓ 4a21e3da ✓✗✓ 4c3d4a41 ✗✓✓ 4c416de3 ✗✗✗ 4c7dc4dd ✗✗✗ 4e34c42c ✗✗✗ 53fb4810 ✓✓✓ 5545f144 ✗✗✗ 581f7754 ✗✗✗ 58490d8a ✓✓✓ 58f5dbd5 ✓✓✓ 5961cc34 ✓✓✓ 5dbc8537 ✓✗✗ 62593bfd ✗✗✗ 64efde09 ✗✓✗ 65b59efc ✓✓✓ 67e490f4 ✓✓✓ 6e453dd6 ✓✓✓ 6e4f6532 ✓✗✗ 6ffbe589 ✗✗✗ 71e489b6 ✓✓✗ 7491f3cf ✓✗✓ 7666fa5d ✓✓✗ 78332cb0 ✗✗✗ 7b0280bc ✓✗✗ 7b3084d4 ✓✓✗ 7b5033c1 ✗✗✓ 7b80bb43 ✗✗✗ 7c66cb00 ✗✗✓ 7ed72f31 ✓✓✗ 800d221b ✗✗✗ 80a900e0 ✓✓✓ 8698868d ✗✗✗ 88bcf3b4 ✗✗✗ 88e364bc ✗✗✗ 89565ca0 ✓✗✓ 898e7135 ✓✓✓ 8b7bacbf ✗✗✗ 8b9c3697 ✓✗✗ 8e5c0c38 ✓✗✓ 8f215267 ✓✓✓ 8f3a5a89 ✓✓✓ 9385bd28 ✓✓✗ 97d7923e ✓✗✓ 981571dc ✓✓✓ 9aaea919 ✓✓✓ 9bbf930d ✗✗✗ a251c730 ✓✓✓ a25697e4 ✗✗✗ a32d8b75 ✗✗✗ a395ee82 ✓✓✓ a47bf94d ✗✓✓ a6f40cea ✓✓✓ aa4ec2a5 ✓✓✓ abc82100 ✗✗✗ b0039139 ✓✗✓ b10624e5 ✓✓✓ b5ca7ac4 ✓✓✓ b6f77b65 ✓✓✗ b99e7126 ✓✓✓ b9e38dc0 ✗✗✗ bf45cf4b ✓✓✓ c4d067a0 ✓✓✓ c7f57c3e ✓✗✗ cb2d8a2c ✓✓✗ cbebaa4b ✓✓✗ d35bdbdc ✗✗✗ d59b0160 ✓✓✓ d8e07eb2 ✓✓✓ da515329 ✗✗✗ db0c5428 ✓✓✓ db695cfb ✓✓✓ dbff022c ✓✓✗ dd6b8c4b ✗✗✗ de809cff ✗✗✗ dfadab01 ✗✓✗ e12f9a14 ✗✗✗ e3721c99 ✗✗✗ e376de54 ✓✓✗ e8686506 ✓✓✓ e87109e9 ✓✓✗ edb79dae ✗✓✓ eee78d87 ✓✗✗ f560132c ✗✗✗ f931b4a8 ✓✓✓ faa9f03d ✗✗✗ fc7cae8d ✗✓✗ ARC-AGI-1 Public Eval 400 tasks Task Max High Low 00576224 ✓✓✓ 009d5c81 ✓✓✓ 00dbd492 ✓✓✓ 03560426 ✓✓✓ 05a7bcf2 ✓✓✓ 0607ce86 ✓✓✓ 0692e18c ✓✓✓ 070dd51e ✓✓✓ 08573cc6 ✓✓✓ 0934a4d8 ✓✓✓ 09c534e7 ✓✓✗ 0a1d4ef5 ✓✓✓ 0a2355a6 ✓✓✓ 0b17323b ✓✓✓ 0bb8deee ✓✓✓ 0becf7df ✓✓✓ 0c786b71 ✓✓✓ 0c9aba6e ✓✓✓ 0d87d2a6 ✗✗✗ 0e671a1a ✓✓✓ 0f63c0b9 ✓✓✓ 103eff5b ✓✓✓ 11e1fe23 ✓✓✓ 12422b43 ✓✓✓ 12997ef3 ✓✓✓ 12eac192 ✓✓✓ 136b0064 ✓✓✓ 13713586 ✓✓✓ 137f0df0 ✓✓✓ 140c817e ✓✓✓ 14754a24 ✓✗✗ 15113be4 ✓✓✓ 15663ba9 ✓✓✓ 15696249 ✓✓✓ 16b78196 ✓✗✗ 17b80ad2 ✓✓✓ 17cae0c1 ✓✓✓ 18419cfa ✓✓✓ 184a9768 ✓✓✓ 195ba7dc ✓✓✓ 1990f7a8 ✓✓✓ 19bb5feb ✓✓✓ 1a2e2828 ✓✓✓ 1a6449f1 ✓✓✓ 1acc24af ✗✗✗ 1c02dbbe ✓✓✓ 1c0d0a4b ✓✓✓ 1c56ad9f ✓✓✓ 1d0a4b61 ✓✓✓ 1d398264 ✓✓✓ 1da012fc ✓✓✓ 1e81d6f9 ✓✓✓ 1e97544e ✓✓✓ 2037f2c7 ✓✓✓ 2072aba6 ✓✓✓ 20818e16 ✓✓✓ 20981f0e ✓✓✓ 212895b5 ✗✗✗ 21f83797 ✓✓✓ 22a4bbc2 ✓✓✓ 25094a63 ✗✗✗ 2546ccf6 ✓✓✓ 256b0a75 ✗✓✓ 2685904e ✓✓✓ 2697da3f ✓✓✓ 2753e76c ✓✓✓ 27a77e38 ✓✓✓ 27f8ce4f ✓✓✓ 281123b4 ✓✓✓ 292dd178 ✓✓✓ 29700607 ✓✓✓ 2a5f8217 ✓✓✓ 2b01abd0 ✓✓✓ 2c0b0aff ✓✓✓ 2c737e39 ✓✓✓ 2f0c5170 ✓✓✓ 310f3251 ✓✓✓ 3194b014 ✓✓✓ 319f2597 ✓✓✓ 31adaf00 ✗✗✓ 31d5ba1a ✓✓✓ 32e9702f ✓✓✓ 332efdb3 ✓✓✓ 3391f8c0 ✓✓✓ 33b52de3 ✓✓✓ 3490cc26 ✓✓✓ 34b99a2b ✓✓✓ 351d6448 ✓✓✓ 358ba94e ✓✓✓ 37d3e8b2 ✓✓✓ 3979b1a8 ✓✓✓ 3a301edc ✓✓✓ 3b4c2228 ✓✓✓ 3d31c5b3 ✓✓✓ 3ed85e70 ✓✓✓ 3ee1011a ✓✓✓ 3f23242b ✓✓✓ 40f6cd08 ✓✓✓ 414297c0 ✓✓✓ 423a55dc ✓✓✓ 42918530 ✓✓✓ 42a15761 ✓✓✓ 4364c1c4 ✓✓✓ 456873bc ✓✓✓ 45737921 ✓✓✓ 45bbe264 ✓✓✓ 477d2879 ✓✗✗ 47996f11 ✗✗✗ 48131b3c ✓✓✓ 4852f2fa ✓✓✓ 48f8583b ✓✓✓ 4aab4007 ✓✓✓ 4acc7107 ✓✓✓ 4b6b68e5 ✓✗✓ 4c177718 ✓✓✓ 4cd1b7b2 ✓✓✓ 4e45f183 ✓✓✓ 4e469f39 ✓✓✓ 4f537728 ✓✓✓ 4ff4c9da ✗✗✓ 505fff84 ✓✓✓ 506d28a5 ✓✓✓ 50a16a69 ✓✓✓ 50aad11f ✓✓✓ 50f325b5 ✗✗✓ 516b51b7 ✓✓✓ 5207a7b5 ✓✓✓ 5289ad53 ✓✓✓ 52fd389e ✓✓✓ 54db823b ✓✓✓ 55059096 ✓✓✓ 551d5bf1 ✓✓✓ 55783887 ✓✓✓ 575b1a71 ✓✓✓ 5783df64 ✓✓✓ 5833af48 ✓✓✓ 58743b76 ✓✓✓ 58e15b12 ✓✓✓ 59341089 ✓✓✓ 5a5a2103 ✓✓✓ 5af49b42 ✓✓✓ 5b526a93 ✓✓✓ 5b692c0f ✓✓✓ 5b6cbef5 ✓✓✓ 5d2a5c43 ✓✓✓ 5ffb2104 ✓✓✓ 604001fa ✓✓✓ 60a26a3e ✓✓✓ 60c09cac ✓✓✓ 626c0bcc ✓✓✓ 62ab2642 ✓✓✗ 62b74c02 ✓✓✓ 639f5a19 ✓✓✓ 642248e4 ✓✓✓ 642d658d ✓✓✓ 64a7c07e ✓✓✓ 66e6c45b ✓✓✓ 66f2d22f ✓✓✓ 67636eac ✓✓✓ 67b4a34d ✓✓✓ 67c52801 ✓✓✓ 68b67ca3 ✓✓✓ 692cd3b6 ✓✓✓ 695367ec ✓✓✓ 696d4842 ✓✓✓ 69889d6e ✓✓✓ 6a11f6da ✓✓✓ 6ad5bdfd ✓✓✓ 6df30ad6 ✓✓✓ 6ea4a07e ✓✓✓ 6f473927 ✓✓✓ 7039b2d7 ✓✓✓ 705a3229 ✓✓✓ 712bf12e ✓✓✓ 72207abc ✓✓✓ 72a961c9 ✓✓✓ 73182012 ✓✓✓ 73c3b0d8 ✓✓✓ 73ccf9c2 ✗✗✗ 759f3fd3 ✓✓✓ 762cd429 ✓✓✓ 770cc55f ✓✓✓ 782b5218 ✓✓✓ 79369cc6 ✓✓✓ 7953d61e ✓✓✓ 79fb03f4 ✓✗✗ 7bb29440 ✓✓✓ 7c8af763 ✓✓✓ 7c9b52a0 ✓✓✓ 7d18a6fb ✓✓✓ 7d1f7ee8 ✓✓✓ 7d419a02 ✓✓✓ 7e02026e ✓✓✓ 7ee1c6ea ✓✓✓ 817e6c09 ✓✓✓ 81c0276b ✓✓✓ 833dafe3 ✓✓✓ 845d6e51 ✓✓✓ 84db8fc4 ✓✓✓ 84f2aca1 ✓✓✓ 8597cfd7 ✓✓✓ 85b81ff1 ✓✓✓ 85fa5666 ✓✗✓ 8719f442 ✓✓✓ 88207623 ✓✓✓ 891232d6 ✗✓✗ 896d5239 ✓✓✗ 8a371977 ✓✓✓ 8b28cd80 ✗✗✗ 8ba14f53 ✓✓✓ 8cb8642d ✓✓✓ 8dae5dfc ✓✓✓ 8e2edd66 ✓✓✓ 8ee62060 ✓✓✓ 8fbca751 ✗✗✗ 90347967 ✓✓✓ 903d1b4a ✓✓✓ 9110e3c5 ✓✓✓ 917bccba ✓✓✓ 929ab4e9 ✓✓✓ 92e50de0 ✗✓✓ 9356391f ✗✓✓ 93b4f4b3 ✓✓✓ 93c31fbe ✓✓✓ 94133066 ✓✓✓ 94414823 ✓✓✓ 94be5b80 ✓✓✓ 95a58926 ✓✓✓ 963f59bc ✓✓✓ 96a8c0cd ✓✓✓ 97239e3d ✓✗✓ 9772c176 ✓✓✓ 981571dc ✓✓✓ 992798f6 ✓✓✓ 99306f82 ✓✓✓ 9a4bb226 ✓✓✓ 9b2a60aa ✓✓✓ 9b365c51 ✓✓✓ 9b4c17c4 ✓✓✓ 9bebae7a ✓✓✓ 9c1e755f ✓✓✓ 9c56f360 ✓✓✓ 9caba7c3 ✓✓✓ 9ddd00f0 ✓✓✓ 9def23fe ✓✓✓ 9f27f097 ✓✓✓ a04b2602 ✓✓✓ a096bf4d ✓✓✓ a3f84088 ✓✓✓ a406ac07 ✓✓✓ a57f2f04 ✓✓✓ a59b95c0 ✓✓✓ a680ac02 ✓✓✓ a8610ef7 ✓✓✗ a934301b ✓✓✓ aa18de87 ✓✓✓ aa300dc3 ✓✓✓ aa4ec2a5 ✓✗✓ aab50785 ✓✓✓ ac0c5833 ✓✗✓ ac2e8ecf ✓✗✓ ac3e2b04 ✓✗✓ ac605cbb ✓✓✓ ad7e01d0 ✓✓✓ ae58858e ✓✓✓ aee291af ✓✓✓ af22c60d ✓✓✓ af24b4cc ✓✓✓ b0722778 ✓✓✓ b0f4d537 ✓✓✓ b15fca0b ✓✓✓ b1fc8b8e ✓✓✓ b20f7c8b ✓✓✗ b457fec5 ✓✓✓ b4a43f3b ✓✓✓ b7999b51 ✓✓✓ b7cb93ac ✓✓✓ b7f8a4d8 ✓✓✓ b7fb29bc ✓✓✓ b942fd60 ✓✗✓ b9630600 ✓✓✗ ba9d41b8 ✓✓✓ baf41dbf ✓✓✓ bb52a14b ✓✓✓ bbb1b8b6 ✓✓✓ bc4146bd ✓✓✓ bcb3040b ✓✓✓ bd14c3bf ✓✓✓ be03b35f ✓✓✓ bf32578f ✓✓✗ bf699163 ✓✓✓ bf89d739 ✓✓✓ c074846d ✓✓✓ c1990cce ✓✓✓ c3202e5a ✓✓✓ c35c1b4c ✓✓✓ c48954c1 ✓✓✓ c62e2108 ✓✓✓ c64f1187 ✓✓✓ c658a4bd ✓✓✓ c663677b ✓✓✓ c6e1b8da ✓✓✗ c7d4e6ad ✓✓✓ c87289bb ✓✓✓ c8b7cc0f ✓✓✓ c92b942c ✓✓✓ c97c0139 ✓✓✓ ca8de6ea ✓✓✓ ca8f78db ✓✓✓ cad67732 ✓✓✓ cb227835 ✓✓✓ ccd554ac ✓✓✓ cd3c21df ✓✓✓ ce039d91 ✓✓✓ ce8d95cc ✓✓✓ cf133acc ✓✓✓ cfb2ce5a ✓✓✓ d017b73f ✗✗✗ d19f7514 ✓✓✓ d282b262 ✓✓✓ d2acf2cb ✓✓✓ d304284e ✓✓✓ d37a1ef5 ✓✓✓ d47aa2ff ✓✓✓ d492a647 ✓✓✓ d4b1c2b1 ✓✓✓ d4c90558 ✓✓✓ d56f2372 ✓✓✓ d5c634a2 ✓✓✓ d931c21c ✓✓✗ d94c3b52 ✓✓✓ da2b0fe3 ✓✓✓ da515329 ✗✗✗ dc2aa30b ✓✓✓ dc2e9a9d ✓✓✓ dd2401ed ✗✓✗ de493100 ✓✗✓ df8cc377 ✓✓✓ e0fb7511 ✓✓✓ e133d23d ✓✓✓ e1baa8a4 ✓✓✓ e1d2900e ✓✗✓ e2092e0c ✓✓✓ e21a174a ✓✓✓ e345f17b ✓✓✓ e4075551 ✓✓✓ e41c6fd3 ✓✓✓ e57337a4 ✓✓✓ e5790162 ✓✓✓ e5c44e8f ✓✓✓ e619ca6e ✓✓✗ e633a9e5 ✓✓✓ e66aafb8 ✓✓✓ e681b708 ✓✓✓ e69241bd ✓✓✓ e6de6e8f ✗✗✓ e74e1818 ✓✓✓ e760a62e ✓✓✓ e7639916 ✓✓✓ e78887d1 ✓✓✓ e7a25a18 ✓✓✓ e7b06bea ✓✓✓ e7dd8335 ✓✓✓ e872b94a ✓✓✓ e88171ec ✓✓✓ e95e3d8e ✓✓✓ e99362f0 ✓✓✓ e9ac8c9e ✓✓✓ e9b4f6fc ✓✓✓ e9bb6954 ✓✓✓ e9c9d9a1 ✓✓✓ ea959feb ✓✓✓ ea9794b1 ✓✓✓ ecaa0ec1 ✓✓✓ ed74f2f2 ✓✓✓ ed98d772 ✓✓✓ ef26cbf6 ✓✓✓ f0afb749 ✓✓✓ f0df5ff0 ✓✓✓ f21745ec ✓✓✓ f3b10344 ✗✗✗ f3cdc58f ✓✓✓ f3e62deb ✓✓✓ f4081712 ✓✓✓ f45f5ca7 ✓✓✓ f5aa3634 ✓✓✓ f5c89df1 ✓✓✓ f823c43c ✓✓✓ f83cb3f6 ✓✓✓ f8be4b64 ✓✓✓ f9a67cb5 ✓✓✗ f9d67f8b ✗✗✓ fafd9572 ✓✓✓ fb791726 ✓✓✓ fc754716 ✓✓✓ fd096ab6 ✓✓✓ fd4b2b02 ✓✓✓ fe9372f3 ✓✓✓ fea12743 ✓✓✓ ff72ca3e ✓✓✓