Search-SWE

MiMo-V2.6-Flash

reasoning effort high · Last Updated

xiaomi/mimo-v2.6-flash

Task results

Task catalog ↗
Task results for MiMo-V2.6-Flash
TaskHardwareAgentProviderTrajectory
agentic-search CPUCodex (v0.164.0)OpenRouter11.71h 54min31,136,61993,41630,654,688$0.18Trace →
agentic-search-vectorless CPUCodex (v0.164.0)OpenRouter0.045min1,396,039405,674698,023$0.21Trace →
biomedical-graph-retrieval CPUCodex (v0.164.0)OpenRouter0.014min2,723,34628,3292,314,624$0.07Trace →
budgeted-bimetric-retrieval CPUCodex (v0.164.0)OpenRouter0.018min9,687,61940,1959,430,336$0.07Trace →
code-context-retrieval CPUCodex (v0.164.0)OpenRouter1.654min15,242,97055,22014,164,672$0.21Trace →
evolving-personal-memory CPUCodex (v0.164.0)OpenRouter8.231min4,616,822311,5203,666,760$0.23Trace →
filtered-ann-retrieval CPUCodex (v0.164.0)OpenRouter0.048min13,615,175168,51712,797,536$0.20Trace →
financial-report-qa CPUCodex (v0.164.0)OpenRouter6.950min9,258,45442,8768,808,960$0.10Trace →
icsi-meeting-memory CPUCodex (v0.164.0)OpenRouter0.057min12,708,85879,08512,530,816$0.08Trace →
incremental-vector-search CPUCodex (v0.164.0)OpenRouter0.71h 59min84,985,689152,79984,499,712$0.35Trace →
join-aware-table-retrieval CPUCodex (v0.164.0)OpenRouter20.050min14,599,29068,87814,375,936$0.09Trace →
large-scale-formula-retrieval CPUCodex (v0.164.0)OpenRouter0.024min5,012,09847,9594,635,008$0.08Trace →
long-document-reranking CPUCodex (v0.164.0)OpenRouter0.07min4,031,943108,0533,435,264$0.12Trace →
long-pdf-evidence CPUCodex (v0.164.0)OpenRouter0.057min15,085,91871,13014,762,816$0.11Trace →
long-pdf-evidence-vectorless CPUCodex (v0.164.0)OpenRouter0.056min14,690,336106,37714,490,112$0.10Trace →
memory-constrained-dense-retrieval CPUCodex (v0.164.0)OpenRouter0.020min4,519,13259,6044,021,280$0.10Trace →
narrative-clue-search CPUCodex (v0.164.0)OpenRouter8.221min3,223,959103,0502,674,304$0.11Trace →
negative-constraint-retrieval CPUCodex (v0.164.0)OpenRouter20.01h 25min14,224,750459,54810,656,315$0.66Trace →
paper-metadata-retrieval CPUCodex (v0.164.0)OpenRouter0.033min8,763,56642,9448,402,816$0.09Trace →
reasoning-query-rewriting CPUCodex (v0.164.0)OpenRouter10.02h 0min27,023,822621,37819,763,424$1.25Trace →
rust-bm25-search CPUCodex (v0.164.0)OpenRouter0.017min6,010,97550,3465,732,736$0.07Trace →
scientific-paper-qa CPUCodex (v0.164.0)OpenRouter20.02h 0min16,093,014592,36011,906,816$0.79Trace →
sparse-retrieval-pruning CPUCodex (v0.164.0)OpenRouter0.055min3,900,004532,9042,453,150$0.36Trace →
visual-document-qa CPUCodex (v0.164.0)OpenRouter6.823min6,722,31531,1526,460,608$0.06Trace →
budgeted-candidate-reranking GPUCodex (v0.164.0)OpenRouter0.014min4,662,96725,6514,449,152$0.05Trace →
chunqiu-temporal-retrieval GPUCodex (v0.164.0)OpenRouter0.025min4,497,97842,5594,202,688$0.07Trace →
code-embedding-finetuning GPUCodex (v0.164.0)OpenRouter0.01h 30min2,968,279548,063696,721$0.47Trace →
code-reranker-finetuning GPUCodex (v0.164.0)OpenRouter25.01h 17min15,602,013126,13512,112,320$0.56Trace →
composed-image-search GPUCodex (v0.164.0)OpenRouter0.012min1,540,08047,0931,252,928$0.06Trace →
constrained-molecular-search GPUCodex (v0.164.0)OpenRouter8.111min2,470,19250,1462,084,288$0.07Trace →
decision-model-training GPUCodex (v0.164.0)OpenRouter0.039min5,307,87551,7684,997,440$0.07Trace →
document-page-search GPUCodex (v0.164.0)OpenRouter3.91h 0min15,378,49468,93014,976,581$0.12Trace →
instruction-following-embedding GPUCodex (v0.164.0)OpenRouter0.054min29,272,393247,83726,020,672$0.60Trace →
khmer-embedding-training GPUCodex (v0.164.0)OpenRouter17.71h 14min5,432,77051,3565,250,688$0.05Trace →
layout-page-search GPUCodex (v0.164.0)OpenRouter0.013min1,626,76454,2191,312,768$0.06Trace →
legal-stance-retrieval GPUCodex (v0.164.0)OpenRouter0.025min6,753,62858,9125,787,168$0.17Trace →
lightweight-query-retrieval GPUCodex (v0.164.0)OpenRouter0.051min8,469,69639,3227,640,288$0.15Trace →
methodology-inspiration-retrieval GPUCodex (v0.164.0)OpenRouter0.032min8,005,80146,1257,370,976$0.12Trace →
multilingual-vector-postprocessing GPUCodex (v0.164.0)OpenRouter20.123min5,413,80433,3534,773,184$0.11Trace →
protein-text-bidirectional-retrieval GPUCodex (v0.164.0)OpenRouter0.045min5,324,62127,5794,135,150$0.19Trace →
quantity-aware-retrieval GPUCodex (v0.164.0)OpenRouter0.09min1,714,59417,8471,460,576$0.04Trace →
query-encoder-alignment GPUCodex (v0.164.0)OpenRouter0.01min256,9744,129164,672$0.01Trace →
same-place-image-search GPUCodex (v0.164.0)OpenRouter0.08min743,89611,141698,880$0.01Trace →
sketch-text-search GPUCodex (v0.164.0)OpenRouter1.41h 52min20,011,47488,44319,133,632$0.20Trace →
spoken-query-search GPUCodex (v0.164.0)OpenRouter2.02h 0min25,674,079534,30022,962,322$0.59Trace →
training-error-tracing GPUCodex (v0.164.0)OpenRouter20.040min18,824,90556,47817,373,891$0.27Trace →
video-moment-search GPUCodex (v0.164.0)OpenRouter10.916min2,668,85130,2502,585,152$0.03Trace →

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).