Traces

Every run: the agent's full trajectory, its submissions over the 20-hour budget, and the final score.

Kolmogorov Audio Compression

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.0369$28.7191.1M330k19h 38m3432
2GLM-5.30.0349$48.94118.7M352k19h 6m3842
3GLM-5.30.0298$35.1891.1M311k19h 41m36613
4GLM-5.30.0264$29.0175.1M335k17h 41m3116
5GLM-5.30.0243$29.3996.9M323k19h 43m3561

Astronomy Toolkit

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.6985$72.52246.9M707k19h 16m5996
2GLM-5.30.4028$57.26195.8M552k12h 7m4268
3GLM-5.30.4017$72.89226.5M514k17h 29m4766
4GLM-5.30.3930$77.97220.3M559k15h 22m4812
5GLM-5.30.2893$64.65196.5M549k19h 1m4198

Cranelift Codegen Optimization

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.0162$93.99346.3M474k11h 28m7662
2GLM-5.30.0129$109.37311.0M484k17h 22m7709
3GLM-5.30.0102$60.68223.7M392k7h 14m6154
4GLM-5.30.0000$110.08353.7M515k18h 43m76118
5GLM-5.30.0000$69.79258.7M392k7h 49m7037

Dart Style in Haskell

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.5299$165.74599.8M1.1M19h 6m139851
2GLM-5.30.4696$158.77595.9M645k19h 33m117712
3GLM-5.30.4653$144.84516.1M611k19h 39m106622
4GLM-5.30.4595$150.71560.0M706k19h 32m126335
5GLM-5.30.3788$98.82359.7M600k20h 0m7381

FFmpeg libswscale Optimization

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.1752$226.40818.4M2.0M19h 33m188418
2GLM-5.30.1410$148.35524.2M1.7M16h 26m13038
3GLM-5.30.1369$139.76505.8M1.2M11h 14m10979
4GLM-5.30.1231$175.55619.5M1.7M16h 47m15098
5GLM-5.30.0783$177.95625.3M1.7M17h 1m14208

Crash-Proof Flash Filesystem

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.0173$171.41627.5M1.2M19h 32m138514
2GLM-5.30.0173$236.98874.4M1.4M19h 51m190831
3GLM-5.30.0111$148.21526.6M1.0M18h 49m117516
4GLM-5.30.0049$108.31386.0M1.3M16h 54m8515
5GLM-5.30.0039$154.89566.6M1.0M19h 45m135018

FrogsGame Post-Training

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.8380$21.9547.4M206k19h 31m2332
2GLM-5.30.7220$12.9144.6M176k19h 17m2472
3GLM-5.30.3120$16.4057.5M195k19h 21m3142
4GLM-5.30.1640$20.8241.0M222k19h 4m2143
5GLM-5.30.1580$16.5958.5M233k19h 38m2651

Stepper Music Sequencer GBA

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.6108$356.801301.6M3.0M19h 1m267119
2GLM-5.30.4715$281.501008.4M2.3M16h 6m226416
3GLM-5.30.4046$276.191010.3M1.8M15h 31m207021
4GLM-5.30.3427$275.261018.0M1.7M17h 39m209039
5GLM-5.30.1440$63.57232.5M498k3h 14m4876

Git to Zig

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.1556$225.36842.0M964k19h 45m201217
2GLM-5.30.1531$195.54692.6M689k19h 54m14802
3GLM-5.30.1390$212.77708.2M852k19h 49m164049
4GLM-5.30.1378$248.79932.5M906k19h 55m221921
5GLM-5.30.1219$188.86712.0M621k19h 43m131816

Granite Mamba2 Inference Optimization

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.8927$85.79297.0M636k6h 7m6347
2GLM-5.30.8719$51.39174.8M618k4h 8m3828
3GLM-5.30.8401$45.93158.7M655k5h 28m3436
4GLM-5.30.7528$38.94126.7M486k4h 14m3515
5GLM-5.30.7402$85.05310.6M673k18h 16m6458

Higgs Uncertainty Inference

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.1632$24.1184.1M350k8h 48m3108
2GLM-5.30.1005$24.4671.1M367k17h 31m2837
3GLM-5.30.0517$9.9130.6M263k2h 10m1675
4GLM-5.30.0375$14.9449.5M308k3h 28m2232
5GLM-5.30.0251$17.6860.0M356k3h 29m2463

SGLang Inference System Optimization

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.0534$68.77241.2M448k15h 36m6508
2GLM-5.30.0211$67.37244.9M383k14h 17m69820
3GLM-5.30.0004$46.58170.2M369k10h 37m5917
4GLM-5.30.0000$50.83180.8M386k12h 52m5764
5GLM-5.30.0000$68.85206.5M390k14h 23m60717

Lean 4 Kernel Type Checker in Pascal

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.8650$139.71481.8M637k18h 8m8596
2GLM-5.30.8372$111.12401.3M867k19h 42m90410
3GLM-5.30.4065$133.87385.6M660k19h 35m6773
4GLM-5.30.1056$134.85474.9M964k19h 7m11053
5GLM-5.30.0000$113.59400.7M1.0M18h 31m9625

libexpat Optimization

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.2372$181.79673.8M1.1M8h 14m15013
2GLM-5.30.2303$183.58670.1M1.1M9h 50m146018
3GLM-5.30.1823$327.581209.3M1.8M18h 23m263546
4GLM-5.30.1583$236.28865.5M1.6M18h 53m199134
5GLM-5.30.0000$216.67774.3M1.6M16h 7m175419

Lua Native Compiler

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.9976$201.34735.2M1.4M11h 37m16603
2GLM-5.30.9953$142.13511.6M1.2M9h 22m11782
3GLM-5.30.9599$274.761000.1M1.8M19h 32m22735
4GLM-5.30.9527$236.30883.7M1.1M17h 20m18357
5GLM-5.30.0000$251.10909.6M1.3M19h 55m19186

Machine-Learned Interatomic Potential

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.9901$21.3550.8M160k19h 5m3531
2GLM-5.30.9615$15.6237.8M156k19h 29m2942
3GLM-5.30.9184$31.3162.0M243k18h 46m3072
4GLM-5.30.8163$20.3255.4M207k19h 30m3502
5GLM-5.30.7858$25.2849.5M200k18h 31m2852

Medium-Range Weather Forecast

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.5873$22.6171.9M228k18h 20m3142
2GLM-5.30.4818$19.7547.5M201k18h 1m2735
3GLM-5.30.4745$28.0396.3M327k19h 20m3637
4GLM-5.30.4723$21.0347.8M206k18h 36m2932
5GLM-5.30.4679$26.3077.9M252k19h 29m3642

MEG Speech Decoding

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.2244$22.6851.0M188k18h 6m3152
2GLM-5.30.2036$34.9674.2M244k14h 55m3722
3GLM-5.30.1492$27.3490.5M318k15h 42m3713
4GLM-5.30.1032$33.8077.7M262k19h 7m3532
5GLM-5.30.0000$49.90134.3M340k19h 4m46921

Wan 2.1 on MAX/Mojo

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.9144$51.16175.7M465k14h 14m4612
2GLM-5.30.9094$42.10140.4M401k16h 39m4042
3GLM-5.30.9087$43.47155.3M374k16h 47m4782
4GLM-5.30.0945$60.08134.7M351k20h 1m4321
5GLM-5.30.0239$77.39173.5M511k19h 46m46411

MS/MS De Novo Generation

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.1050$34.1375.8M305k18h 42m3362
2GLM-5.30.1036$36.9494.3M365k18h 38m3593
3GLM-5.30.0941$44.44143.9M417k17h 20m4508
4GLM-5.30.0000$28.5571.5M295k19h 12m3322
5GLM-5.30.0000$43.34125.4M382k18h 56m4632

Multi-GPU Efficient Finetuning

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.2857$12.8826.8M144k19h 0m2342
2GLM-5.30.2857$10.4218.4M127k19h 10m1756
3GLM-5.30.0714$17.6133.2M221k18h 56m2112
4GLM-5.30.0000$20.2634.9M208k19h 31m2114
5GLM-5.30.0000$17.6732.5M199k19h 29m2152

Notebook Compression

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.2770$25.8686.5M327k19h 49m3703
2GLM-5.30.2744$34.96116.4M391k19h 2m4352
3GLM-5.30.2089$39.2889.8M362k20h 0m3691
4GLM-5.30.2058$59.65209.0M565k15h 11m5488
5GLM-5.30.1261$28.9386.9M409k13h 16m3282

Flight-Sim Renderer in OpenGL

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.0166$91.30316.9M852k19h 35m8917
2GLM-5.30.0054$111.06362.5M755k19h 29m89310
3GLM-5.30.0038$97.66297.0M573k19h 4m6335
4GLM-5.30.0006$94.18300.9M621k18h 57m59210
5GLM-5.30.0000$103.09375.1M554k19h 46m79012

Optimizer Design

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.2349$19.9835.0M224k17h 41m1923
2GLM-5.30.2192$17.3736.9M195k19h 6m2412
3GLM-5.30.2082$16.4826.8M198k19h 23m1843
4GLM-5.30.0984$17.0155.5M261k19h 4m2775
5GLM-5.30.0965$13.4723.0M159k18h 59m1802

PostgreSQL 18 on SQLite

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.1593$191.86711.5M862k19h 53m174617
2GLM-5.30.1484$202.20744.7M1.0M19h 42m17897
3GLM-5.30.1374$201.89706.7M843k19h 32m15859
4GLM-5.30.1374$155.77570.5M772k19h 52m13842
5GLM-5.30.0989$219.76822.5M966k19h 50m190311

Quantum ESPRESSO pw.x in Rust

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.1707$145.94507.1M1.3M19h 30m108213
2GLM-5.30.1382$132.96476.8M1.3M19h 42m98412
3GLM-5.30.1301$108.12382.4M1.3M19h 32m8382
4GLM-5.30.0650$123.53403.7M1.6M19h 19m95711
5GLM-5.30.0000$103.68348.9M1.2M19h 23m79514

Qubit Routing

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.7535$23.4664.7M277k19h 3m2676
2GLM-5.30.7343$45.4890.2M303k16h 27m3654
3GLM-5.30.6907$44.40135.6M383k15h 8m39212
4GLM-5.30.6370$18.5955.0M231k19h 13m2873
5GLM-5.30.6071$32.20106.0M372k12h 23m3755

Reconnaissance Blind Chess Recovery

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.5357$48.03138.0M410k14h 33m3889
2GLM-5.30.4214$29.5279.9M302k14h 26m3056
3GLM-5.30.3607$36.0573.5M339k13h 30m2594
4GLM-5.30.2893$25.2683.8M284k14h 38m3394
5GLM-5.30.2571$27.0269.8M315k14h 43m2536

Fitness-Recap Video in Remotion

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.1894$78.50212.7M442k18h 20m5306
2GLM-5.30.1828$113.14407.8M483k19h 2m8049
3GLM-5.30.1426$118.44380.0M784k18h 55m98618
4GLM-5.30.1053$110.52388.9M908k15h 36m9619
5GLM-5.30.0837$90.93307.2M485k16h 10m6368

Snooker Prediction

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.1976$71.12251.5M765k18h 56m65415
2GLM-5.30.1013$57.78201.7M793k14h 16m53113
3GLM-5.30.0590$53.37163.6M430k17h 18m3736
4GLM-5.30.0311$89.97246.3M519k19h 15m49513
5GLM-5.30.0000$94.15290.7M465k18h 52m62711

SPICE Circuit Simulator in Rust

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.5354$171.96618.6M1.7M18h 15m14414
2GLM-5.30.4949$235.96867.9M1.6M17h 11m190518
3GLM-5.30.4444$158.88578.0M1.3M10h 53m11554
4GLM-5.30.4141$190.24693.2M1.5M19h 41m170530
5GLM-5.30.3131$152.26545.3M1.3M16h 57m116423

Synthetic Music Diarization

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.4816$35.34125.4M361k17h 30m46211
2GLM-5.30.4802$31.6279.0M303k18h 9m3611
3GLM-5.30.4602$19.7741.7M196k19h 9m2534
4GLM-5.30.4285$28.0377.4M258k18h 22m3669
5GLM-5.30.4265$27.2778.5M240k19h 37m3746

Vision-only TORCS Racing Bot

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.0490$52.08118.9M330k18h 59m3542
2GLM-5.30.0406$79.08158.6M318k18h 36m4068
3GLM-5.30.0363$101.89234.7M397k19h 55m4871
4GLM-5.30.0278$99.55248.0M367k18h 54m56314
5GLM-5.30.0180$78.52153.1M353k18h 42m3969

Verilog Simulator in Swift

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GLM-5.30.8236$236.70879.4M1.3M20h 1m192411
2GLM-5.30.8151$195.25722.0M1.1M19h 28m170610
3GLM-5.30.7757$229.93817.6M1.1M19h 47m180019
4GLM-5.30.7723$247.89875.5M1.2M19h 55m180417
5GLM-5.30.7551$285.371061.6M1.4M19h 46m236723