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
1Grok 4.60.0221$127.10131.5M617k19h 58m61730
2Grok 4.60.0192$168.00174.0M498k19h 58m76868
3Grok 4.60.0171$114.81118.3M558k19h 40m59947
4Grok 4.60.0126$162.44156.4M577k19h 37m63461
5Grok 4.60.0106$206.03201.8M768k19h 59m84563

Astronomy Toolkit

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.5299$15.5715.6M323k1h 52m1064
2Grok 4.60.3993$48.1146.2M395k3h 31m2086
3Grok 4.60.2000$34.6834.5M354k2h 42m1758
4Grok 4.60.1971$39.6635.3M400k3h 43m16511
5Grok 4.60.0000$23.0923.9M317k4h 4m1523

Cranelift Codegen Optimization

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.0000$113.60103.4M918k9h 35m4266
2Grok 4.60.0000$240.25234.7M1.3M16h 47m89524
3Grok 4.60.0000$138.74131.9M1.2M8h 47m5876
4Grok 4.60.0000$138.58127.2M927k7h 29m4889
5Grok 4.60.0000$108.30103.9M740k6h 37m4476

Dart Style in Haskell

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.5864$513.86512.7M2.7M20h 0m205459
2Grok 4.60.3015$389.00386.7M2.2M19h 53m164542
3Grok 4.60.1785$547.31552.2M2.4M19h 52m217774
4Grok 4.60.0000$485.05493.3M2.2M19h 52m1992109
5Grok 4.60.0000$402.09394.7M2.1M17h 23m152478

FFmpeg libswscale Optimization

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.0360$419.25404.2M2.4M19h 56m1517114
2Grok 4.60.0000$1121.421126.0M2.4M18h 11m3693434
3Grok 4.60.0000$68.4059.6M603k3h 50m2294
4Grok 4.60.0000$132.56129.2M778k5h 28m56227
5Grok 4.60.0000$380.75368.6M1.8M11h 52m137067

Crash-Proof Flash Filesystem

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.0610$858.44859.7M2.4M19h 46m2990501
2Grok 4.60.0241$445.27424.9M2.8M19h 59m1572200
3Grok 4.60.0117$262.00255.9M1.7M13h 11m103920
4Grok 4.60.0046$267.41252.9M1.7M10h 33m927132
5Grok 4.60.0031$181.59175.3M1.4M8h 20m69324

FrogsGame Post-Training

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.2460$18.5719.8M283k19h 59m1521
2Grok 4.60.2420$9.6812.9M262k19h 22m1322
3Grok 4.60.1280$47.6745.1M287k18h 35m2595
4Grok 4.60.0640$49.0243.5M397k19h 5m2214
5Grok 4.60.0340$48.2946.2M363k16h 43m2375

Stepper Music Sequencer GBA

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.5671$305.84301.9M2.2M11h 0m122745
2Grok 4.60.3420$868.26872.5M3.4M16h 30m3114373
3Grok 4.60.2607$223.96223.8M1.5M7h 20m88218
4Grok 4.60.2416$205.23195.4M1.9M7h 34m79228
5Grok 4.60.2197$129.50120.8M2.0M7h 52m54019

Git to Zig

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.1693$652.70657.8M2.2M19h 59m2502122
2Grok 4.60.1618$614.44610.4M2.6M18h 0m232473
3Grok 4.60.1516$806.68819.7M2.2M19h 59m2981186
4Grok 4.60.1352$482.88483.0M1.6M11h 22m176784
5Grok 4.60.1269$408.61405.7M1.8M19h 55m155848

Granite Mamba2 Inference Optimization

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.7305$58.7452.8M626k4h 20m20613
2Grok 4.60.7245$46.6739.7M692k3h 57m16910
3Grok 4.60.6586$55.7047.5M700k4h 6m19410
4Grok 4.60.6473$57.8851.1M725k4h 13m2105
5Grok 4.60.2777$27.6423.0M551k3h 18m1272

Higgs Uncertainty Inference

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.9143$37.2334.8M407k3h 22m15810
2Grok 4.60.8991$42.7739.2M529k3h 38m1839
3Grok 4.60.8827$39.9237.7M462k3h 38m1818
4Grok 4.60.8188$274.70264.2M1.9M19h 38m107867
5Grok 4.60.2104$116.88112.5M824k9h 43m47052

SGLang Inference System Optimization

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.0322$488.14454.2M2.3M19h 33m170383
2Grok 4.60.0025$162.38158.0M937k10h 27m64914
3Grok 4.60.0012$313.02296.5M1.9M18h 41m112019
4Grok 4.60.0007$96.0091.3M481k3h 53m3293
5Grok 4.60.0000$419.76405.9M2.2M19h 4m158660

Lean 4 Kernel Type Checker in Pascal

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.3824$105.13105.0M1.0M6h 34m4902
2Grok 4.60.0000$114.14110.5M1.1M7h 47m4924
3Grok 4.60.0000$407.54406.7M2.0M19h 47m1527137
4Grok 4.60.0000$250.10240.1M1.8M13h 13m97043
5Grok 4.60.0000$233.84217.8M2.5M19h 53m92930

libexpat Optimization

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.1479$150.76148.7M804k4h 11m58026
2Grok 4.60.1393$108.10108.1M736k4h 5m43511
3Grok 4.60.1022$79.4275.8M516k2h 56m2737
4Grok 4.60.0000$93.7799.5M677k3h 26m43510
5Grok 4.60.0000$71.7870.3M399k1h 58m2775

Lua Native Compiler

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.6384$176.45167.1M1.5M7h 26m6538
2Grok 4.60.0000$110.36107.0M986k5h 11m4478
3Grok 4.60.0000$271.55264.4M2.0M9h 49m106131
4Grok 4.60.0000$676.54663.0M4.1M19h 59m2429277
5Grok 4.60.0000$250.76240.6M2.1M11h 13m93016

Machine-Learned Interatomic Potential

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.8964$27.9418.3M258k19h 42m1472
2Grok 4.60.8799$71.4258.5M227k19h 56m3079
3Grok 4.60.8342$80.8855.9M251k19h 37m26510
4Grok 4.60.7267$28.2320.6M259k19h 44m1593
5Grok 4.60.5686$61.5946.3M291k18h 29m2504

Medium-Range Weather Forecast

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.5804$120.40104.0M444k19h 13m46437
2Grok 4.60.5522$208.02184.0M911k19h 55m82653
3Grok 4.60.5495$238.57230.6M626k20h 0m86162
4Grok 4.60.4530$43.0735.9M245k19h 35m2251
5Grok 4.60.4154$92.4866.7M263k19h 27m3036

MEG Speech Decoding

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.3041$128.46122.4M984k15h 5m54523
2Grok 4.60.2265$217.69204.9M1.2M19h 25m87823
3Grok 4.60.2107$132.86129.3M708k19h 25m57133
4Grok 4.60.1844$95.1683.3M284k18h 58m36211
5Grok 4.60.1245$71.3867.2M365k9h 53m29310

Wan 2.1 on MAX/Mojo

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.9085$149.23138.7M1.3M12h 2m5862
2Grok 4.60.9047$29.2330.9M270k3h 12m1581
3Grok 4.60.9010$76.5166.4M639k9h 42m2732
4Grok 4.60.2716$127.41108.5M1.7M19h 30m4644
5Grok 4.60.2296$214.90198.0M1.7M19h 41m75913

MS/MS De Novo Generation

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.1023$289.71280.9M1.3M20h 0m110991
2Grok 4.60.0709$86.3377.3M437k10h 44m3176
3Grok 4.60.0000$51.2439.8M426k15h 47m2435
4Grok 4.60.0000$192.75185.2M1.2M14h 10m76664
5Grok 4.60.0000$86.8176.6M468k16h 53m3369

Multi-GPU Efficient Finetuning

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.0000$47.0350.4M373k19h 22m26113
2Grok 4.60.0000$18.5324.3M327k17h 14m1947
3Grok 4.60.0000$32.8828.8M249k18h 12m1965
4Grok 4.60.0000$23.9518.0M271k19h 15m1454
5Grok 4.60.0000$20.0718.4M217k18h 24m1568

Notebook Compression

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.2286$85.3184.1M498k7h 48m33114
2Grok 4.60.2146$36.9834.6M412k13h 20m2205
3Grok 4.60.2098$96.3888.3M636k11h 39m37712
4Grok 4.60.2060$71.5475.2M388k6h 34m38127
5Grok 4.60.2019$87.9283.6M470k8h 37m36110

Flight-Sim Renderer in OpenGL

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.0526$317.62274.4M1.8M19h 55m1078110
2Grok 4.60.0382$286.38260.8M1.4M19h 57m984103
3Grok 4.60.0273$488.84437.7M2.0M19h 56m1536307
4Grok 4.60.0266$454.46403.4M2.0M19h 58m143575
5Grok 4.60.0147$328.36283.8M1.8M19h 57m108457

Optimizer Design

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.2557$99.1498.0M624k19h 20m42418
2Grok 4.60.1957$53.9066.7M503k19h 24m39524
3Grok 4.60.1570$59.5357.5M295k19h 26m25811
4Grok 4.60.1368$54.4949.3M462k19h 16m24010
5Grok 4.60.0839$80.4575.7M333k19h 42m2793

PostgreSQL 18 on SQLite

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.1593$509.89496.5M1.6M19h 56m169670
2Grok 4.60.1538$461.37465.7M1.9M17h 59m184249
3Grok 4.60.0989$257.56255.7M1.5M8h 44m95217
4Grok 4.60.0714$398.74383.4M2.7M19h 51m145734
5Grok 4.60.0495$148.44141.0M985k6h 17m55910

Quantum ESPRESSO pw.x in Rust

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.3984$196.95181.0M1.6M13h 51m68114
2Grok 4.60.3008$282.67273.1M1.6M13h 49m97640
3Grok 4.60.2683$1211.081191.6M2.2M19h 59m3559243
4Grok 4.60.1057$123.16111.5M1.7M10h 28m42913
5Grok 4.60.0976$96.1488.3M1.1M7h 19m36012

Qubit Routing

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.7350$123.13116.9M502k9h 53m41428
2Grok 4.60.6883$68.0360.3M544k3h 28m2496
3Grok 4.60.6689$82.9075.9M477k3h 6m28317
4Grok 4.60.6619$46.4748.9M429k4h 33m2747
5Grok 4.60.5924$84.0182.0M496k4h 20m33610

Reconnaissance Blind Chess Recovery

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.6036$87.2970.2M604k8h 33m2599
2Grok 4.60.5857$104.5199.3M819k14h 17m45324
3Grok 4.60.4071$107.3799.7M852k13h 36m44317
4Grok 4.60.2786$78.3465.1M979k10h 29m3018
5Grok 4.60.0929$172.71159.1M1.1M14h 32m60914

Fitness-Recap Video in Remotion

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.0338$1762.941204.0M2.8M20h 0m4175289
2Grok 4.60.0279$646.10557.5M1.5M19h 48m2043149
3Grok 4.60.0245$555.19490.3M1.5M19h 56m202090
4Grok 4.60.0242$1149.83792.5M1.8M19h 58m2836250
5Grok 4.60.0216$746.94531.5M1.4M19h 55m1941163

Snooker Prediction

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.3789$466.11391.2M2.0M19h 43m148873
2Grok 4.60.3048$106.1286.0M725k5h 58m3785
3Grok 4.60.1779$108.4587.9M851k6h 23m34415
4Grok 4.60.1175$125.27104.6M768k11h 27m4469
5Grok 4.60.0252$77.1660.3M816k4h 29m27113

SPICE Circuit Simulator in Rust

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.5657$717.25714.4M2.7M18h 22m2483266
2Grok 4.60.5253$835.65836.5M3.1M20h 0m3128205
3Grok 4.60.4848$586.10582.0M3.0M17h 34m232289
4Grok 4.60.4646$1191.201187.2M2.5M19h 58m387996
5Grok 4.60.4545$165.93154.1M860k5h 37m56337

Synthetic Music Diarization

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.4661$133.75133.0M655k19h 52m57842
2Grok 4.60.4323$188.19192.5M1.1M19h 45m87156
3Grok 4.60.4120$228.84232.3M993k19h 44m91863
4Grok 4.60.3985$299.53304.5M1.6M19h 27m128154
5Grok 4.60.0000$501.54509.4M1.9M18h 17m193796

Vision-only TORCS Racing Bot

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.1206$114.8193.3M882k19h 41m42610
2Grok 4.60.0335$207.50166.7M1.1M18h 40m69223
3Grok 4.60.0122$159.04128.5M975k18h 46m56920
4Grok 4.60.0043$218.28183.9M1.3M14h 37m75215
5Grok 4.60.0000$57.5052.5M468k7h 59m2523

† score adjusted by manual review; the trace's Judge tab carries the measured score and the rationale.

Verilog Simulator in Swift

task page →5 runs · indexed 2026-09-02
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Grok 4.60.8253$223.65211.2M1.9M12h 53m88431
2Grok 4.60.8116$347.28325.2M2.2M19h 55m122256
3Grok 4.60.7517$518.10508.9M2.5M19h 53m184250
4Grok 4.60.6901$168.65160.9M1.6M9h 35m69125
5Grok 4.60.5188$107.0399.0M943k6h 25m40219