Traces

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

Astronomy Toolkit

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.2056$84.96112.8M364k5h 13m42819
2Muse Spark 1.30.1015$57.0570.7M278k5h 16m36111
3Muse Spark 1.30.0976$81.1381.9M258k12h 38m3972
4Muse Spark 1.30.0279$94.7498.7M274k19h 49m42638
5Muse Spark 1.30.0028$18.8927.3M239k2h 57m1754

Cranelift Codegen Optimization

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.0231$411.74433.6M529k18h 32m8754
2Muse Spark 1.30.0192$411.77452.6M517k18h 56m9448
3Muse Spark 1.30.0168$495.72581.2M737k19h 59m15303
4Muse Spark 1.30.0000$472.59577.9M275k18h 58m127612
5Muse Spark 1.30.0000$394.96454.8M322k15h 45m102112

Crash-Proof Flash Filesystem

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.0177$523.70619.4M1.1M20h 0m14091
2Muse Spark 1.30.0173$532.50628.8M1.0M20h 0m15521
3Muse Spark 1.30.0002$572.11638.9M1.2M19h 59m13841
4Muse Spark 1.30.0000$486.57544.5M845k19h 59m14211
5Muse Spark 1.30.0000$536.44601.3M1.3M19h 59m13241

Dart Style in Haskell

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.5513$525.73756.2M1.1M19h 57m16378
2Muse Spark 1.30.5258$522.21631.2M1.1M19h 59m17575
3Muse Spark 1.30.5020$463.61501.9M1.1M19h 20m12659
4Muse Spark 1.30.3680$506.26577.2M925k19h 53m15671
5Muse Spark 1.30.0010$582.91707.6M1.1M19h 56m174010

FFmpeg libswscale Optimization

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.1277$499.70617.7M951k19h 17m140124
2Muse Spark 1.30.1028$438.57557.6M1.0M16h 30m134023
3Muse Spark 1.30.1003$186.76239.3M502k7h 59m55914
4Muse Spark 1.30.0996$157.06200.2M465k6h 34m50412
5Muse Spark 1.30.0779$279.58336.6M595k9h 26m68817

Fitness-Recap Video in Remotion

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.1827$179.291045.0M1.1M19h 56m22908
2Muse Spark 1.30.1338$375.85806.9M1.0M20h 0m19273
3Muse Spark 1.30.0362$213.98610.8M874k19h 50m16363
4Muse Spark 1.30.0269$429.57453.8M790k19h 52m12452
5Muse Spark 1.30.0263$363.14496.5M863k19h 59m13371

Flight-Sim Renderer in OpenGL

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.1055$206.91689.2M886k10h 35m16811
2Muse Spark 1.30.0410$299.15649.2M932k19h 37m15882
3Muse Spark 1.30.0386$233.75628.1M517k18h 2m12825
4Muse Spark 1.30.0254$147.68673.6M1.0M19h 47m15618
5Muse Spark 1.30.0250$351.72511.0M807k19h 59m13001

FrogsGame Post-Training

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.6080$42.7636.7M222k19h 41m2412
2Muse Spark 1.30.1340$12.4916.7M201k19h 20m1793
3Muse Spark 1.30.0580$11.2613.8M164k19h 42m1633
4Muse Spark 1.30.0380$29.7531.6M256k19h 58m2611
5Muse Spark 1.30.0320$20.0329.7M151k19h 22m2411

Git to Zig

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.1427$402.31656.1M1.1M18h 56m16323
2Muse Spark 1.30.1274$518.45593.9M920k19h 54m14652
3Muse Spark 1.30.1062$484.87561.8M975k20h 0m14991
4Muse Spark 1.30.1006$482.98557.2M1.0M20h 0m159415
5Muse Spark 1.30.0000$641.52782.0M1.4M19h 59m15801

Granite Mamba2 Inference Optimization

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.8136$31.6238.2M374k3h 49m1865
2Muse Spark 1.30.8036$155.17181.2M487k18h 42m54511
3Muse Spark 1.30.7809$183.67203.2M648k10h 29m55711
4Muse Spark 1.30.6457$320.20329.1M590k13h 47m77312
5Muse Spark 1.30.6302$29.9288.8M460k3h 13m3259

Higgs Uncertainty Inference

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.8206$37.7359.2M320k3h 38m30012
2Muse Spark 1.30.7474$18.6743.3M220k4h 23m2506
3Muse Spark 1.30.3726$625.66665.8M679k19h 55m1128223
4Muse Spark 1.30.3641$40.7252.7M332k3h 49m2578
5Muse Spark 1.30.0300$39.3251.4M265k3h 54m2618

Kolmogorov Audio Compression

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.0247$73.5580.1M351k10h 48m3927
2Muse Spark 1.30.0215$70.2670.8M222k11h 3m36612
3Muse Spark 1.30.0200$64.2165.3M198k19h 54m4045
4Muse Spark 1.30.0192$39.49134.3M306k19h 56m5683
5Muse Spark 1.30.0124$64.7658.2M224k18h 13m38411

Lean 4 Kernel Type Checker in Pascal

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.9776$411.77410.4M713k19h 52m7992
2Muse Spark 1.30.7421$154.98818.9M1.3M19h 49m18052
3Muse Spark 1.30.5609$411.55416.2M856k19h 58m9591
4Muse Spark 1.30.5029$384.24384.5M940k19h 50m9371
5Muse Spark 1.30.3530$389.61408.6M860k19h 59m10151

libexpat Optimization

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.0996$565.41641.8M1.1M19h 58m142216
2Muse Spark 1.30.0752$635.38703.6M636k19h 52m13423
3Muse Spark 1.30.0491$376.252119.7M1.7M19h 47m4639144
4Muse Spark 1.30.0092$609.32704.0M1.1M19h 57m14972
5Muse Spark 1.30.0000$571.21716.3M974k19h 59m16832

Lua Native Compiler

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.9976$451.94515.6M941k16h 46m12612
2Muse Spark 1.30.4969$541.57625.6M1.2M20h 0m14471
3Muse Spark 1.30.4002$540.38602.2M792k20h 0m14001
4Muse Spark 1.30.0000†$75.32109.8M406k5h 42m4616
5Muse Spark 1.30.0000$637.70724.8M869k20h 0m16311

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

Machine-Learned Interatomic Potential

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.9755$18.5916.6M98k19h 46m2193
2Muse Spark 1.30.9671$61.9258.4M240k19h 36m4104
3Muse Spark 1.30.9529$5.4710.6M104k19h 34m1772
4Muse Spark 1.30.8349$17.6515.6M188k19h 8m1741
5Muse Spark 1.30.2516$58.4551.9M245k19h 22m3575

Medium-Range Weather Forecast

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.4646$17.4032.8M150k19h 29m25011
2Muse Spark 1.30.4252$39.9763.7M234k18h 47m40110
3Muse Spark 1.30.4057$47.0341.9M228k13h 41m31111
4Muse Spark 1.30.3909$44.6041.6M175k17h 28m2837
5Muse Spark 1.30.3160$38.6381.9M260k18h 49m43413

MEG Speech Decoding

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.3657$29.6328.5M123k19h 14m29012
2Muse Spark 1.30.1430$33.9830.2M124k18h 44m2784
3Muse Spark 1.30.1334$8.5030.2M106k18h 10m3177
4Muse Spark 1.30.1187$59.1456.9M180k14h 50m3673
5Muse Spark 1.30.0000$32.8629.1M146k18h 59m2696

MS/MS De Novo Generation

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.1200$62.0457.0M274k16h 49m3293
2Muse Spark 1.30.1131$41.9045.4M252k18h 50m3047
3Muse Spark 1.30.1082$55.0349.7M225k19h 54m3611
4Muse Spark 1.30.1039$52.8046.7M249k19h 50m3516
5Muse Spark 1.30.0000$54.2546.3M268k18h 59m3102

Multi-GPU Efficient Finetuning

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.3095$25.5021.9M121k19h 28m1972
2Muse Spark 1.30.1429$16.1814.5M230k20h 0m1591
3Muse Spark 1.30.0000$22.9620.6M169k19h 58m1575
4Muse Spark 1.30.0000$12.9112.2M103k19h 52m1421
5Muse Spark 1.30.0000$30.0827.4M216k19h 49m2041

Notebook Compression

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.2729$60.3265.3M317k6h 41m35711
2Muse Spark 1.30.2285$27.0179.4M314k19h 7m4078
3Muse Spark 1.30.2112$78.1476.4M260k11h 8m3583
4Muse Spark 1.30.1309$65.0467.7M249k11h 11m3716
5Muse Spark 1.30.1170$63.5077.9M313k5h 49m3263

Optimizer Design

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.1949$17.6714.5M143k19h 59m1501
2Muse Spark 1.30.1110$29.2125.7M102k19h 53m3072
3Muse Spark 1.30.1023$23.8020.6M126k19h 41m2602
4Muse Spark 1.30.0000$14.0411.9M125k19h 29m1791
5Muse Spark 1.30.0000$22.0318.0M123k19h 58m2012

PostgreSQL 18 on SQLite

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.3407$551.80666.0M898k19h 58m17144
2Muse Spark 1.30.1099$305.431604.1M2.2M20h 0m36902
3Muse Spark 1.30.0549$522.84598.7M1.2M20h 0m15251
4Muse Spark 1.30.0549$518.76603.9M1.1M19h 59m15721
5Muse Spark 1.30.0495$525.95562.0M986k20h 0m13871

Quantum ESPRESSO pw.x in Rust

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.4228$418.71442.8M1.1M20h 0m11191
2Muse Spark 1.30.4228$514.22583.5M1.3M19h 59m14151
3Muse Spark 1.30.3577$352.28387.2M1.4M19h 43m11533
4Muse Spark 1.30.1463$336.34348.3M1.1M19h 35m9399
5Muse Spark 1.30.0163$435.17464.8M1.2M19h 59m11961

Qubit Routing

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.6821$28.9785.7M430k5h 33m39621
2Muse Spark 1.30.6752$165.04162.2M432k18h 15m70625
3Muse Spark 1.30.6538$51.5157.6M507k5h 41m28610
4Muse Spark 1.30.6245$65.1569.0M497k6h 34m38113
5Muse Spark 1.30.6205$141.97139.6M348k19h 5m58013

Reconnaissance Blind Chess Recovery

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.4500$83.30135.3M1.0M14h 48m4594
2Muse Spark 1.30.1107$38.26178.6M566k18h 5m65436
3Muse Spark 1.30.0286$195.94191.4M452k19h 7m6847
4Muse Spark 1.30.0107$34.91148.6M508k14h 14m57715
5Muse Spark 1.30.0036$74.56160.5M728k11h 20m62810

SGLang Inference System Optimization

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.1159$180.28204.8M347k10h 27m67014
2Muse Spark 1.30.0000$429.35454.7M706k19h 28m11658
3Muse Spark 1.30.0000$58.3660.2M197k8h 23m3863
4Muse Spark 1.30.0000$52.36191.1M288k7h 43m6805
5Muse Spark 1.30.0000$73.5999.0M263k5h 24m4593

Snooker Prediction

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.3826$317.27411.4M767k19h 50m10324
2Muse Spark 1.30.3439$236.90672.0M957k19h 49m15695
3Muse Spark 1.30.0370$445.62445.7M615k19h 19m8578
4Muse Spark 1.30.0000$518.18574.1M1.1M19h 44m12906
5Muse Spark 1.30.0000$449.88491.2M832k19h 52m11762

SPICE Circuit Simulator in Rust

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.5758$268.721376.2M2.2M19h 51m33282
2Muse Spark 1.30.5657$460.65531.6M1.1M17h 48m13474
3Muse Spark 1.30.5455$559.10664.1M1.1M19h 47m15405
4Muse Spark 1.30.4646$547.17608.4M1.3M19h 44m13313
5Muse Spark 1.30.3939$557.80621.9M1.5M19h 58m13911

Stepper Music Sequencer GBA

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.5524$214.491195.2M2.0M12h 19m28461
2Muse Spark 1.30.3727$330.621484.1M2.3M19h 59m330530
3Muse Spark 1.30.3146$507.64559.8M1.1M19h 23m13542
4Muse Spark 1.30.0797$528.52575.6M1.2M19h 59m13131
5Muse Spark 1.30.0480$556.48567.3M735k19h 51m11075

Synthetic Music Diarization

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.4569$29.57105.7M282k17h 55m5198
2Muse Spark 1.30.4340$126.99121.2M301k15h 7m4379
3Muse Spark 1.30.4243$152.31160.5M376k14h 14m5007
4Muse Spark 1.30.4206$121.45118.8M279k16h 52m51021
5Muse Spark 1.30.3962$92.9691.1M258k17h 50m45410

Verilog Simulator in Swift

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.5582$533.21614.5M1.1M19h 59m15771
2Muse Spark 1.30.5497$512.38619.2M1.1M19h 59m15981
3Muse Spark 1.30.4932$275.581496.0M1.9M19h 49m33613
4Muse Spark 1.30.4932$515.26603.1M1.3M19h 58m14732
5Muse Spark 1.30.4384$522.03583.4M1.2M19h 58m14091

Vision-only TORCS Racing Bot

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.0388$56.56261.0M256k19h 53m8465
2Muse Spark 1.30.0144$191.47315.7M497k17h 52m8149
3Muse Spark 1.30.0029$53.07279.6M311k19h 59m98511
4Muse Spark 1.30.0000†$28.0494.3M152k18h 59m4695
5Muse Spark 1.30.0000†$90.6082.5M232k18h 53m4103

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

Wan 2.1 on MAX/Mojo

task page →5 runs · indexed 2026-10-08
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1Muse Spark 1.30.9348$88.2491.7M303k11h 17m4092
2Muse Spark 1.30.9005$242.82232.9M452k19h 26m6281
3Muse Spark 1.30.4425$115.32111.6M449k19h 59m3851
4Muse Spark 1.30.4417$126.25155.2M591k19h 59m4651
5Muse Spark 1.30.4389$215.78199.3M678k15h 8m50224