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 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra1.0000$2526.291186.0M2.6M19h 52m274860
2GPT-6 Astra1.0000$2608.331211.7M2.6M19h 14m257757
3GPT-6 Astra1.0000$2728.801295.7M2.6M19h 47m292059
4GPT-6 Astra1.0000$1955.97919.3M1.8M13h 34m203127
5GPT-6 Astra0.0696$683.33329.9M756k19h 26m87916

Astronomy Toolkit

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.8045$135.6567.7M278k2h 40m2775
2GPT-6 Astra0.7969$2365.471121.6M2.3M19h 43m254849
3GPT-6 Astra0.7058$233.15111.6M353k2h 41m35610
4GPT-6 Astra0.7013$86.8245.8M224k2h 2m2145
5GPT-6 Astra0.6917$365.75173.3M402k3h 17m4577

Cranelift Codegen Optimization

task page →4 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.3006$2589.931231.7M1.9M19h 38m2587122
2GPT-6 Astra0.2357$2732.071287.7M2.4M19h 34m282997
3GPT-6 Astra0.2194$2585.691232.6M2.0M18h 22m264379
4GPT-6 Astra0.1650$2812.601334.7M2.3M19h 56m293492

Dart Style in Haskell

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.9601$2439.091157.2M2.2M16h 56m267130
2GPT-6 Astra0.9508$2034.34972.8M1.7M15h 13m221642
3GPT-6 Astra0.9472$895.41422.2M827k6h 22m95814
4GPT-6 Astra0.9426$2766.391322.7M2.4M19h 53m298474
5GPT-6 Astra0.8965$171.4681.1M258k2h 1m2777

FFmpeg libswscale Optimization

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.2755$169.8582.4M307k2h 7m2907
2GPT-6 Astra0.2021$313.12153.5M377k2h 54m4589
3GPT-6 Astra0.1983$306.45149.1M351k2h 48m4448
4GPT-6 Astra0.1741$262.36125.2M349k2h 37m3796
5GPT-6 Astra0.1179$275.12132.2M355k2h 35m3966

Crash-Proof Flash Filesystem

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.6090$92.2045.6M190k1h 44m1774
2GPT-6 Astra0.5968$497.70237.1M397k4h 16m51614
3GPT-6 Astra0.5902$154.8475.5M253k1h 42m2595
4GPT-6 Astra0.5835$114.9855.8M198k1h 55m2085
5GPT-6 Astra0.5519$228.77111.2M286k2h 11m3325

FrogsGame Post-Training

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.9700$116.6760.2M205k18h 38m2558
2GPT-6 Astra0.9160$256.74124.3M279k19h 10m36511
3GPT-6 Astra0.9000$49.9727.6M160k19h 17m1586
4GPT-6 Astra0.8840$265.88128.3M310k19h 14m3908
5GPT-6 Astra0.6760$444.71214.4M392k19h 40m52110

Stepper Music Sequencer GBA

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.9974$3129.481515.4M2.4M19h 55m337649
2GPT-6 Astra0.9897$2759.221308.1M2.5M19h 59m294238
3GPT-6 Astra0.9861$2721.061296.9M2.6M19h 56m297826
4GPT-6 Astra0.9793$2755.431303.4M2.5M19h 59m291637
5GPT-6 Astra0.9655$2773.421319.2M2.5M19h 57m294130

Git to Zig

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.9864$3111.971407.9M3.4M19h 58m311948
2GPT-6 Astra0.9049$3006.211380.7M3.3M20h 0m312960
3GPT-6 Astra0.8950$2950.791350.4M3.3M19h 50m303152
4GPT-6 Astra0.8938$3057.891408.3M3.3M19h 59m320496
5GPT-6 Astra0.8893$3214.361493.6M3.2M19h 59m334799

Granite Mamba2 Inference Optimization

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.8707$2626.811259.5M2.3M18h 25m283958
2GPT-6 Astra0.8658$272.49128.6M369k2h 40m3677
3GPT-6 Astra0.8334$251.26124.6M331k2h 35m4094
4GPT-6 Astra0.8266$344.95163.8M406k3h 8m4598
5GPT-6 Astra0.8198$254.82122.3M354k2h 28m3747

Higgs Uncertainty Inference

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra1.0000$164.9780.6M296k2h 47m3068
2GPT-6 Astra1.0000$50.7531.1M211k1h 50m1846
3GPT-6 Astra0.9768$52.9333.2M214k1h 55m2005
4GPT-6 Astra0.4310$132.0368.3M275k2h 40m2895
5GPT-6 Astra0.2020$84.8443.9M239k1h 53m2086

SGLang Inference System Optimization

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.5349$1954.03921.6M1.6M19h 56m193231
2GPT-6 Astra0.0000$2665.441249.7M2.4M19h 33m269942
3GPT-6 Astra0.0000$2441.521153.7M2.1M19h 6m255341
4GPT-6 Astra0.0000$2574.701208.7M2.3M17h 31m262265
5GPT-6 Astra0.0000$1722.33819.7M1.5M19h 18m184442

Lean 4 Kernel Type Checker in Pascal

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.9807$122.0864.3M290k2h 50m2824
2GPT-6 Astra0.9807$62.9837.7M245k1h 38m2035
3GPT-6 Astra0.9807$123.0561.8M300k2h 8m2603
4GPT-6 Astra0.9791$34.1522.2M179k1h 14m1333
5GPT-6 Astra0.9791$174.6285.4M285k2h 50m3114

libexpat Optimization

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.2941$510.01240.5M518k4h 1m5427
2GPT-6 Astra0.2776$415.35191.6M517k3h 32m4446
3GPT-6 Astra0.2744$177.8683.5M336k2h 10m2835
4GPT-6 Astra0.1534$13.196.7M92k53m624
5GPT-6 Astra0.1516$19.3310.3M134k38m783

Lua Native Compiler

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra1.0000$46.1926.4M213k1h 23m1347
2GPT-6 Astra1.0000$35.9723.6M186k1h 21m1334
3GPT-6 Astra1.0000$131.7664.1M279k2h 36m2327
4GPT-6 Astra0.9976$79.5942.6M240k2h 8m1959
5GPT-6 Astra0.9976$93.6948.6M237k1h 56m2066

Machine-Learned Interatomic Potential

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.9884$148.0172.8M245k17h 13m2838
2GPT-6 Astra0.9884$97.1352.7M210k16h 8m2576
3GPT-6 Astra0.9842$127.3365.0M249k17h 10m2877
4GPT-6 Astra0.9702$43.8726.5M154k17h 20m1529
5GPT-6 Astra0.9688$90.0151.8M262k15h 58m2867

Medium-Range Weather Forecast

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.6474$297.67145.6M335k19h 47m44114
2GPT-6 Astra0.6387$27.5517.3M152k18h 9m13512
3GPT-6 Astra0.6362$98.7753.1M216k17h 56m25714
4GPT-6 Astra0.6334$244.67123.1M350k18h 49m44513
5GPT-6 Astra0.0000$488.12238.0M474k17h 46m61016

MEG Speech Decoding

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.3919$899.22430.4M993k17h 31m106415
2GPT-6 Astra0.3729$274.60132.9M344k8h 16m41113
3GPT-6 Astra0.3704$154.2876.5M270k4h 15m2979
4GPT-6 Astra0.3557$452.78217.0M444k15h 46m52714
5GPT-6 Astra0.3431$526.36253.0M673k17h 55m69420

Wan 2.1 on MAX/Mojo

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.9623$128.9866.9M202k2h 32m2523
2GPT-6 Astra0.9217$103.1656.8M189k2h 7m2363
3GPT-6 Astra0.9173$183.3492.0M226k1h 45m3052
4GPT-6 Astra0.9153$76.8942.2M156k1h 3m1873
5GPT-6 Astra0.9128$82.9447.1M162k1h 37m2053

MS/MS De Novo Generation

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.1731$851.54402.0M882k16h 28m92815
2GPT-6 Astra0.1695$1429.90682.5M1.4M13h 9m163115
3GPT-6 Astra0.1667$818.15390.6M876k18h 17m98821
4GPT-6 Astra0.1525$440.96211.9M461k8h 1m5467
5GPT-6 Astra0.1519$408.97200.1M427k5h 57m5745

Multi-GPU Efficient Finetuning

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.2143$155.5978.1M233k19h 27m2996
2GPT-6 Astra0.1905$229.70113.1M288k19h 53m3646
3GPT-6 Astra0.1905$125.1064.1M212k19h 30m2616
4GPT-6 Astra0.0714$291.48141.3M320k19h 38m4026
5GPT-6 Astra0.0000$142.9574.8M253k19h 46m3117

Notebook Compression

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.6122$2035.40958.4M1.9M17h 45m213253
2GPT-6 Astra0.6063$2124.531006.7M2.1M18h 49m230842
3GPT-6 Astra0.5422$1208.04578.6M1.3M19h 17m138924
4GPT-6 Astra0.4322$274.97132.6M360k5h 47m4077
5GPT-6 Astra0.4121$327.78158.4M411k6h 16m48210

Flight-Sim Renderer in OpenGL

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.9965$2541.331210.9M2.5M19h 50m283027
2GPT-6 Astra0.9845$2685.881278.3M2.6M20h 0m296138
3GPT-6 Astra0.9844$2735.711297.4M2.6M19h 42m288429
4GPT-6 Astra0.9841$2589.721228.2M2.6M19h 59m280134
5GPT-6 Astra0.9841$2728.191288.8M2.7M19h 57m294147

Optimizer Design

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.3513$198.61101.4M290k17h 55m3815
2GPT-6 Astra0.1940$334.17159.8M312k18h 23m4127
3GPT-6 Astra0.1777$89.9548.3M203k19h 38m24410
4GPT-6 Astra0.0929$278.29132.5M303k16h 52m3787
5GPT-6 Astra0.0523$53.3535.6M191k19h 46m2186

PostgreSQL 18 on SQLite

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.7308$2982.001383.4M2.9M19h 56m307367
2GPT-6 Astra0.7143$3132.521453.4M2.8M19h 51m314044
3GPT-6 Astra0.6593$3139.161477.1M2.8M19h 52m328073
4GPT-6 Astra0.6484$3119.081452.8M2.8M19h 56m310862
5GPT-6 Astra0.6209$3312.101548.3M2.9M20h 0m325561

Quantum ESPRESSO pw.x in Rust

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.9350$571.91270.6M494k3h 53m5704
2GPT-6 Astra0.9350$540.43256.4M446k3h 37m5444
3GPT-6 Astra0.8699$378.55179.5M449k3h 11m4595
4GPT-6 Astra0.8699$279.05132.3M363k2h 37m3724
5GPT-6 Astra0.8293$299.55142.9M339k2h 28m4024

Qubit Routing

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.9207$412.86201.8M473k7h 46m59512
2GPT-6 Astra0.9186$1381.68662.3M1.4M17h 52m162126
3GPT-6 Astra0.9168$2130.531024.2M1.9M19h 50m238445
4GPT-6 Astra0.9131$2239.251067.1M2.1M19h 57m242954
5GPT-6 Astra0.8800$604.79292.5M576k8h 29m71511

Reconnaissance Blind Chess Recovery

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.9857$244.69120.7M324k13h 51m40810
2GPT-6 Astra0.9429$571.36274.8M479k10h 52m6377
3GPT-6 Astra0.8429$330.64163.8M407k12h 13m5269
4GPT-6 Astra0.8357$771.41367.5M716k14h 57m90312
5GPT-6 Astra0.7536$12.034.8M92k34m693

Fitness-Recap Video in Remotion

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.7901$2673.961271.2M2.5M19h 58m299131
2GPT-6 Astra0.7705$2639.521259.7M2.5M19h 50m296127
3GPT-6 Astra0.7148$2733.411304.4M2.5M19h 53m297729
4GPT-6 Astra0.6873$2846.471353.6M2.5M19h 54m311436
5GPT-6 Astra0.6437$2584.511240.4M2.4M19h 47m292023

Snooker Prediction

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.6319$2653.821245.9M2.6M19h 52m276939
2GPT-6 Astra0.6316$2364.831112.0M2.4M19h 31m253951
3GPT-6 Astra0.6308$1877.69889.9M1.7M16h 9m199731
4GPT-6 Astra0.6304$348.12162.7M418k3h 11m4424
5GPT-6 Astra0.6223$321.27149.5M395k3h 5m4014

SPICE Circuit Simulator in Rust

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.9091$2887.401385.4M2.5M19h 50m3121180
2GPT-6 Astra0.8990$347.32164.1M405k2h 56m4147
3GPT-6 Astra0.8990$599.18284.8M531k4h 2m6167
4GPT-6 Astra0.8889$677.88324.1M558k4h 32m6947
5GPT-6 Astra0.8687$172.7283.5M308k2h 47m2884

Synthetic Music Diarization

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.5899$503.76243.1M455k15h 19m59111
2GPT-6 Astra0.5838$329.84157.1M371k18h 6m42613
3GPT-6 Astra0.5821$102.5750.4M226k17h 2m2148
4GPT-6 Astra0.5755$313.33152.4M361k17h 42m45912
5GPT-6 Astra0.5062$228.20112.0M318k15h 43m37915

Vision-only TORCS Racing Bot

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.4570$770.42373.2M731k19h 37m101312
2GPT-6 Astra0.3866$1781.91865.7M1.5M19h 55m213325
3GPT-6 Astra0.2803$1564.06755.2M1.2M19h 40m185716
4GPT-6 Astra0.1048$823.35394.4M766k19h 59m101814
5GPT-6 Astra0.0879$1737.29826.0M1.5M19h 53m193716

Verilog Simulator in Swift

task page →5 runs · indexed 2026-09-22
#ModelRewardCostIn tokOut tokTimeStepsSubmits
1GPT-6 Astra0.9503$104.5349.0M213k1h 48m1875
2GPT-6 Astra0.9469$329.41155.6M366k2h 26m40926
3GPT-6 Astra0.9452$47.7326.0M162k1h 5m1303
4GPT-6 Astra0.9349$94.8544.6M216k1h 33m1746
5GPT-6 Astra0.9264$71.9536.3M192k1h 35m1626