DeepSeek V4.1 Flash 1017
Task results
Task catalog ↗| Task | Hardware | Agent | Provider | Trajectory | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| agentic-search | CPU | Codex | OpenRouter | 0.0 | 2h 0min | 78,190,365 | 130,916 | 77,850,028 | $0.36 | Trace → |
| agentic-search-vectorless | CPU | Codex | OpenRouter | 0.0 | 2h 0min | 13,714,128 | 97,011 | 10,971,392 | $0.50 | Trace → |
| biomedical-graph-retrieval | CPU | Codex | OpenRouter | 0.0 | 1h 44min | 82,338,985 | 321,572 | 82,067,968 | $0.48 | Trace → |
| budgeted-bimetric-retrieval | CPU | Codex | OpenRouter | 0.5 | 1h 7min | 23,005,761 | 151,965 | 21,580,032 | $0.37 | Trace → |
| code-context-retrieval | CPU | Codex | OpenRouter | 16.7 | 1h 59min | 54,729,695 | 211,435 | 53,845,504 | $0.42 | Trace → |
| evolving-personal-memory | CPU | Codex | OpenRouter | 19.5 | 41min | 56,260,315 | 305,119 | 54,305,408 | $0.64 | Trace → |
| filtered-ann-retrieval | CPU | Codex | OpenRouter | 0.0 | 2h 0min | 31,691,383 | 282,144 | 30,673,024 | $0.41 | Trace → |
| financial-report-qa | CPU | Codex | OpenRouter | 6.9 | 1h 41min | 65,787,581 | 258,600 | 64,804,352 | $0.50 | Trace → |
| icsi-meeting-memory | CPU | Codex | OpenRouter | 11.1 | 2h 0min | 57,698,115 | 322,621 | 57,056,512 | $0.46 | Trace → |
| incremental-vector-search | CPU | Codex | OpenRouter | 7.6 | 2h 0min | 37,882,260 | 288,818 | 34,764,126 | $0.75 | Trace → |
| join-aware-table-retrieval | CPU | Codex | OpenRouter | 30.0 | 1h 2min | 17,658,078 | 147,518 | 16,539,934 | $0.31 | Trace → |
| large-scale-formula-retrieval | CPU | Codex | OpenRouter | 0.0 | 1h 20min | 5,526,257 | 107,066 | 5,419,648 | $0.10 | Trace → |
| long-document-reranking | CPU | Codex | OpenRouter | 18.0 | 2h 0min | 9,965,677 | 108,563 | 9,717,769 | $0.13 | Trace → |
| long-pdf-evidence | CPU | Codex | OpenRouter | 16.7 | 2h 0min | 41,466,104 | 204,499 | 40,936,576 | $0.32 | Trace → |
| long-pdf-evidence-vectorless | CPU | Codex | OpenRouter | 50.0 | 2h 0min | 16,763,261 | 117,451 | 16,569,344 | $0.15 | Trace → |
| memory-constrained-dense-retrieval | CPU | Codex | OpenRouter | 0.0 | 1h 29min | 73,957,570 | 326,168 | 73,099,506 | $0.54 | Trace → |
| narrative-clue-search | CPU | Codex | OpenRouter | 9.1 | 2h 0min | 29,003,400 | 176,650 | 26,874,624 | $0.51 | Trace → |
| negative-constraint-retrieval | CPU | Codex | OpenRouter | 70.0 | 1h 11min | 30,450,951 | 214,842 | 29,446,016 | $0.37 | Trace → |
| paper-metadata-retrieval | CPU | Codex | OpenRouter | 12.5 | 1h 45min | 32,263,145 | 187,632 | 31,792,320 | $0.28 | Trace → |
| reasoning-query-rewriting | CPU | Codex | OpenRouter | 30.0 | 1h 58min | 59,851,572 | 159,219 | 58,909,132 | $0.41 | Trace → |
| rust-bm25-search | CPU | Codex | OpenRouter | 0.0 | 38min | 25,640,463 | 193,323 | 23,552,768 | $0.50 | Trace → |
| scientific-paper-qa | CPU | Codex | OpenRouter | 0.0 | 2h 0min | 80,020,322 | 221,219 | 79,240,576 | $0.49 | Trace → |
| sparse-retrieval-pruning | CPU | Codex | OpenRouter | 0.0 | 1h 59min | 89,879,717 | 326,595 | 88,309,151 | $0.70 | Trace → |
| visual-document-qa | CPU | Codex | OpenRouter | 21.7 | 1h 55min | 43,405,107 | 375,545 | 39,592,146 | $0.92 | Trace → |
| budgeted-candidate-reranking | GPU | Codex | OpenRouter | 17.1 | 2h 0min | 60,927,610 | 237,776 | 60,500,864 | $0.39 | Trace → |
| chunqiu-temporal-retrieval | GPU | Codex | OpenRouter | 0.4 | 1h 33min | 66,484,650 | 228,777 | 65,267,152 | $0.52 | Trace → |
| code-embedding-finetuning | GPU | Codex | OpenRouter | 17.0 | 4h 58min | 74,271,355 | 235,389 | 72,477,442 | $0.63 | Trace → |
| code-reranker-finetuning | GPU | Codex | OpenRouter | 0.0 | 38min | 3,470,080 | 64,228 | 3,200,384 | $0.09 | Trace → |
| composed-image-search | GPU | Codex | OpenRouter | 29.6 | 2h 0min | 15,141,433 | 202,640 | 13,275,468 | $0.44 | Trace → |
| constrained-molecular-search | GPU | Codex | OpenRouter | 13.6 | 4h 17min | 214,341,252 | 306,148 | 210,097,152 | $1.45 | Trace → |
| decision-model-training | GPU | Codex | OpenRouter | 0.0 | 16min | 409,275 | 147,003 | 315,776 | $0.10 | Trace → |
| document-page-search | GPU | Codex | OpenRouter | 0.0 | 2h 0min | 10,434,763 | 133,430 | 9,393,920 | $0.26 | Trace → |
| instruction-following-embedding | GPU | Codex | OpenRouter | 89.3 | 1h 31min | 20,907,536 | 156,626 | 20,721,280 | $0.18 | Trace → |
| khmer-embedding-training | GPU | Codex | OpenRouter | 20.9 | 5h 55min | 55,183,710 | 230,635 | 54,538,752 | $0.40 | Trace → |
| layout-page-search | GPU | Codex | OpenRouter | 18.6 | 2h 14min | 18,318,879 | 272,374 | 17,764,608 | $0.30 | Trace → |
| legal-stance-retrieval | GPU | Codex | OpenRouter | 30.2 | 4h 31min | 72,955,224 | 239,275 | 72,687,744 | $0.40 | Trace → |
| lightweight-query-retrieval | GPU | Codex | OpenRouter | 0.0 | 4h 19min | 265,369,029 | 404,644 | 262,744,802 | $1.42 | Trace → |
| methodology-inspiration-retrieval | GPU | Codex | OpenRouter | 1.4 | 4h 18min | 64,544,579 | 208,140 | 63,879,040 | $0.42 | Trace → |
| multilingual-vector-postprocessing | GPU | Codex | OpenRouter | 100.0 | 2h 4min | 30,772,324 | 205,324 | 30,299,581 | $0.29 | Trace → |
| protein-text-bidirectional-retrieval | GPU | Codex | OpenRouter | 2.5 | 1h 0min | 6,442,099 | 172,705 | 6,052,064 | $0.18 | Trace → |
| quantity-aware-retrieval | GPU | Codex | OpenRouter | 0.0 | 12min | 1,943,459 | 68,324 | 1,843,712 | $0.06 | Trace → |
| query-encoder-alignment | GPU | Codex | OpenRouter | 0.0 | 3h 22min | 166,139,205 | 304,899 | 165,714,320 | $0.74 | Trace → |
| same-place-image-search | GPU | Codex | OpenRouter | 52.0 | 2h 0min | 37,915,500 | 205,066 | 37,245,891 | $0.34 | Trace → |
| sketch-text-search | GPU | Codex | OpenRouter | 0.0 | 6h 0min | 18,601,297 | 155,389 | 18,204,346 | $0.21 | Trace → |
| spoken-query-search | GPU | Codex | OpenRouter | 14.3 | 2h 0min | 45,669,931 | 355,065 | 44,942,065 | $0.46 | Trace → |
| training-error-tracing | GPU | Codex | OpenRouter | 0.0 | 2h 51min | 103,296,442 | 330,365 | 99,769,088 | $1.03 | Trace → |
| video-moment-search | GPU | Codex | OpenRouter | 0.0 | <1min | 65,808 | 7,592 | 33,920 | $0.01 | Trace → |
No matching tasks. Try a different search.
47 selected evaluations · valid zero scores included
Score: task reward × 100. Tokens: coding usage; cached tokens are included in input.
Cost is estimated from recorded agent tokens at official API rates in USD (2026-11-01; DeepSeek off-peak pricing).