Awards
$1,500, $1,000 and $500 in every track.
The first, second and third teams in each of the four tracks win cash awards: $12,000 in total.
NeurIPS 2026 Competition Track
Agenthon 2026
Verifiable AI for quantitative finance.
A four-track competition testing whether AI agents can produce finance outputs that survive automated, leakage-controlled, cheat-resistant evaluation.
Submission
Docker agent
Admissibility
g0-g3 gates
Ranking
Track metrics
We're delighted to welcome NVIDIA, Bloomberg and AllianceBernstein as sponsors of Agenthon 2026. Thank you for your support, for helping us organize the competition, and for collaborating with us on its science and task development. Meet our sponsors
Key dates
Key dates.
Registration runs from 17 August to 12 October 2026. Registration and Development close together on 12 October at 23:59 AoE (13 October, 11:59 UTC), the one deadline for Development uploads on every track. Every valid Development upload made before then is run, even if its run starts after the close.
-
17 Aug – 12 Oct
Registration
Sign up and form your team.
-
28 Aug – 12 Oct
Development
Public practice repositories, iteration, and the validation leaderboard.
-
30 Sep
Call for papers closed
Papers for the NeurIPS workshop were due at 23:59 AoE.
-
13 – 25 Oct
Final + Verification
One submission per entered track is evaluated on sealed held-out units. Uploads close on 25 October at 23:59 AoE; organizers verify the top entries after their runs. Results and awards are published in early November.
-
Sat 12 Dec
Workshop at NeurIPS, Atlanta
The Agenthon workshop in Atlanta, Georgia, with posters from the winning teams and the accepted papers.
The competition closes with the Agenthon workshop at NeurIPS in Atlanta, Georgia, on Saturday 12 December, where the winning teams present posters alongside the accepted papers. All dates are given in the official rules, which govern if anything here is out of step.
Teams
Enter as a team of one to three.
You can compete alone or with up to two teammates. You join one team and stay on it, and that one team can enter as many of the four tracks as it wants.
Registration is per team, and only registered teams can submit solutions or appear on the leaderboard. Your team uses one account on the competition platform, chosen by the team — the FAQ explains what that means for everyone else on the team.
Have a question about entering? Start with the FAQ.
Core question
Can AI agents produce finance answers that can be checked by machine?
Agenthon extends the Alphathon program into its first NeurIPS edition. The competition keeps the four-track structure and hard finance setting, while adding sealed held-out data, automated leakage controls, reproducible reruns, and public leaderboards.
Agents in all four tracks run in sandboxed Docker containers without general internet access. Supported Coding, Forecasting and Explainability submissions can call House Nemotron through a restricted connection. Simulation has no network access. Build required dependencies and permitted artifacts into your image before evaluation.
Four tracks
Same competition spine, four finance problems.
Quant-finance coding agents
Build a Docker agent that solves quantitative finance coding tasks under pytest and financial-invariant checks.
- Verb
- solve
- Metric
- pass@1
- Gate
- pytest + invariants
Reasoning-augmented time series
Forecast future panels using time-series data plus a time-stamped text corpus. The score evaluates the forecast distribution; information uplift is a research aim.
- Verb
- forecast
- Metric
- CRPS composite
- Gate
- as-of cutoff + calibration
Accelerated market simulation
Submit an ABIDES-compatible simulator that is faster while preserving matching-engine semantics and market stylized facts. Development results are practice feedback, not controlled Final timing.
- Verb
- simulate
- Metric
- events/sec
- Gate
- semantic regression
Evidence-grounded prediction
Predict labels, values, or rankings over tabular entities, supported by a frozen evidence corpus. Follow the track guide for prediction, evidence and reasoning evaluation.
- Verb
- analyze
- Metric
- Track 4 composite
- Gate
- faithfulness + embargo
Protocol
Submit, check, score.
Each unit is checked for admissibility and evaluated using its track's scoring rules. Participant failures remain in the evaluation denominator; organizer faults are handled separately. If two Final submissions finish a track with the same ranking score, the tie is broken in favour of the one uploaded earlier.
01
Submit
Upload the toolkit-generated ZIP identifying your agent image and proving team registration.
02
Check
g0-g3 verify integrity, schema, cutoff/resource rules, and domain semantics.
03
Score
Track-specific scoring produces Development feedback. Track 1 uses pass@1 without a confidence interval.
Public/private firewall
Transparent grading, sealed answers.
Each track has a public practice repo and a private sealed exam repo. Public repos include practice tasks, examples, local checks, available baselines and participant docs. Private repos hold evaluation tasks, reference answers, scoring material and audit records.
Public practice
Private exam
Public practice units
Private-test held-out units
Examples, local checks and available baselines
Oracle solutions and final scorer
Manifest and canary safety checks
Canary registry and audit logs
Leaderboard
Competition scores, one track at a time.
Development leaderboards on agenthon.net are public and visible to everyone. Each track uses its own scale. Higher is better for the displayed leaderboard scores; the raw Forecasting loss is converted for this display. Use CodaBench for uploads and processing status. Follow the logged-in website for submission opening status. From 5 October 2026, 00:00 AoE (12:00 UTC), the Track 1 board ranks only runs uploaded after 25 September 2026 at 05:00 UTC, for every team (Track 1 README, rules 8 and 9).
The leaderboard is updated periodically from the competition platform.
T1 Coding
T2 Forecasting
T3 Simulation
T4 Explainability
Search teams
T1 Coding
T1 Coding: teams 1–25 of 121
Rank
Team
Score
1
Saifuddin
0.9070
2
Kaile
0.7674
3
huhudawang
0.5349
3
Pluvia
0.5349
5
Proxima
0.5000
6
Arsen Ibragimov
0.4767
7
RandomGuyHavingFun
0.4651
8
Chirayu
0.4302
9
DeepSucc
0.4070
10
dbuddha
0.3837
11
RizYL
0.3721
12
Yan Su
0.3605
13
Quiet Signal
0.3488
13
Nana Champions
0.3488
15
Proof of Alpha
0.3372
16
MindAlchemist
0.3140
17
BetterCallSaul
0.2791
17
jinfanhahaha
0.2791
17
cache_miss
0.2791
17
Agent007
0.2791
21
abc123
0.2674
21
Four Sigma
0.2674
21
Winner
0.2674
24
eagle
0.2558
25
a-team
0.2326
T1 Coding: teams 26–50 of 121
Rank
Team
Score
25
Cornfield Chase
0.2326
25
Formaggio
0.2326
28
Jin & Pei
0.2209
28
EDB AI
0.2209
28
Autonomous Alpha
0.2209
28
fiftyfifty
0.2209
32
Liqourena
0.2093
32
Tengen Toppa
0.2093
32
CMCCGDYDDICT
0.2093
32
Flowfront
0.2093
32
Pluto
0.2093
37
Anonymous
0.1977
37
in3lab
0.1977
37
APTAPT
0.1977
37
Husky River Trading
0.1977
37
Masato
0.1977
42
haibao
0.1860
42
Verifiable Capital
0.1860
42
always win
0.1860
42
Apex
0.1860
42
VerityLovityFalsityCruelty
0.1860
42
Submartingale
0.1860
42
Make It Make Cents 🙏
0.1860
42
eoo0m
0.1860
50
iluss
0.1744
T1 Coding: teams 51–75 of 121
Rank
Team
Score
50
Salt-Lake-Agent
0.1744
50
Just-in-Time
0.1744
50
BoGo
0.1744
50
Paragon
0.1744
55
Undefined
0.1628
55
TappuKiMKC
0.1628
55
AbangFrabi
0.1628
55
verifiquant
0.1628
55
FAWDA
0.1628
55
FedorAzarov
0.1628
55
Powerhouse
0.1628
55
Nietzsche
0.1628
55
Zeal
0.1628
55
Outsiders
0.1628
55
1991
0.1628
55
Eyyyyyy
0.1628
67
Sean
0.1512
67
Money Miner
0.1512
67
Give me a job
0.1512
70
BIN Han
0.1395
70
gheware
0.1395
70
AgenthonTeam
0.1395
70
JessH
0.1395
70
Agenthon_42
0.1395
70
Warkop PuteraPrakoso
0.1395
T1 Coding: teams 76–100 of 121
Rank
Team
Score
76
3226
0.1279
76
RRRRRR
0.1279
76
3Dogs Nexus
0.1279
76
HTHY
0.1279
76
Duo_tech
0.1279
76
Jade
0.1279
82
MicroCat
0.1163
82
hnhparitosh
0.1163
82
Agent_007
0.1163
82
Jumping Tree Frogs
0.1163
82
Yee Quant
0.1163
82
Quant Science Flow
0.1163
88
TEst
0.1047
88
The Mockingjay
0.1047
90
UNIVERSEPLAYER
0.0930
90
lingsio
0.0930
90
Sir Thaddeus
0.0930
90
Banana Kingdom
0.0930
94
Orrery
0.0814
94
Echoing
0.0814
94
Agent Alpha
0.0814
94
Luca
0.0814
94
IMAnonymous
0.0814
99
The Unscheduled Penguins
0.0698
99
quant-coder
0.0698
T1 Coding: teams 101–121 of 121
Rank
Team
Score
101
Team Solo
0.0581
102
team
0.0465
102
GAOKUO
0.0465
104
attention is qkv
0.0349
105
volatility_labs_stony_brook
0.0233
105
Wealth-Building Lab
0.0233
107
Urithiru
0.0116
107
nogugu
0.0116
107
WoldyWei Solo
0.0116
—
Aamir Abdul Azeez
— not scored
—
AgentReady
— not scored
—
DeltaNLP
— not scored
—
AlbertQuant
— not scored
—
Flying Penguin
— not scored
—
9:00pm
— not scored
—
C_cake
— not scored
—
phtree
— not scored
—
DaoyuShu4094
— not scored
—
Future Ability
— not scored
—
Artifical General Invesment
— not scored
—
tokenmaxxing
— not scored
No team matches that search.
T2 Forecasting
T2 Forecasting: teams 1–25 of 154
Rank
Team
Score
1
huhudawang
-0.7554
2
DuML
-0.8317
3
Yan Su
-1.3403
4
alpha-extractor
-1.5401
5
Ywin
-1.5483
6
Flowfront
-1.5500
7
Paragon
-1.5813
8
Made in Heaven
-1.6351
9
HBIT
-1.6656
10
Samoyed
-1.6713
11
iluss
-1.6752
12
RRRRRR
-1.6881
13
Pawsibble
-1.7077
14
Chirayu
-1.7134
15
TheAxiom
-1.7487
16
F1_1
-1.7666
17
Not Today
-1.7697
18
Quiet Signal
-1.7917
19
Cornfield Chase
-1.7953
20
Jin & Pei
-1.8424
21
jinfanhahaha
-1.8436
22
Just-in-Time
-1.8461
23
Proof of Alpha
-1.8515
24
labubu666
-1.8693
25
CMCCGDYDDICT
-1.8758
T2 Forecasting: teams 26–50 of 154
Rank
Team
Score
26
VerityLovityFalsityCruelty
-1.9167
27
Saifuddin
-1.9190
28
5 o'clock
-1.9603
29
Apex
-1.9618
30
HackStreet Boys
-1.9722
31
Nexus of f(x)
-2.0134
32
Warkop PuteraPrakoso
-2.0303
33
agi1
-2.0579
34
LOOOONG
-2.0811
35
Super Debugging
-2.0971
36
Winner
-2.0983
37
DKYnumber1
-2.1028
38
Autonomous Alpha
-2.1097
39
Jigglyduck
-2.1111
40
Hatenabase
-2.1126
41
Alpha Decay
-2.1177
42
TradeFlare
-2.1314
43
1991
-2.1359
44
fiftyfifty
-2.1571
45
Gamblr
-2.1667
46
phtree
-2.1718
47
iKun-XinglinForrest
-2.1770
48
spidy
-2.1786
49
Quant Science Flow
-2.2106
50
Sapien
-2.2129
T2 Forecasting: teams 51–75 of 154
Rank
Team
Score
51
urayaha
-2.2155
52
lifeisluck
-2.2172
53
EDB AI
-2.2191
54
LastDigitsOfPi
-2.2338
55
Masato
-2.2448
56
HG Market System
-2.2556
57
quack
-2.2893
58
Money Miner
-2.2898
59
Sitadel Insecurities
-2.2962
60
JJ LAB
-2.2970
61
Richard Zhu
-2.3207
62
Formaggio
-2.3310
63
pluuuuuus
-2.3551
64
abc123
-2.3567
65
time
-2.3694
66
pranshu rastogi
-2.3774
67
AsOf Quant
-2.3990
68
JessH
-2.4001
69
Agenthon_42
-2.4052
70
Tengen Toppa
-2.4107
71
Bayes Street
-2.4108
72
Javier Amo
-2.4170
73
Yongchang
-2.4173
74
Zeal
-2.4202
75
OOBs
-2.4212
T2 Forecasting: teams 76–100 of 154
Rank
Team
Score
76
The Risk Museum
-2.4274
77
Auror
-2.4301
78
QIE DOG
-2.4329
79
Dog parents
-2.4467
80
lingsio
-2.4480
81
hnhparitosh
-2.4484
82
C_cake
-2.4513
83
jusep
-2.4542
84
in3lab
-2.4570
85
LIZARD
-2.4605
86
SOLO
-2.4613
87
Orion26
-2.4616
88
LiLaiLai
-2.4637
89
tokenmaxxing
-2.4657
90
randomwalk
-2.4666
91
DoublePai
-2.4681
92
cache_miss
-2.4681
93
FAWDA
-2.4720
94
Aamir Abdul Azeez
-2.4722
95
EBal
-2.4734
96
ZZKK
-2.4795
97
In Sync
-2.4848
98
Orrery
-2.4875
99
Natpaphon
-2.4891
100
Verifiable Capital
-2.4918
T2 Forecasting: teams 101–125 of 154
Rank
Team
Score
101
GAOKUO
-2.4932
102
Slipstream
-2.4935
103
SMTM
-2.4938
104
krnb
-2.4961
105
IPL_KKR
-2.5003
106
1111
-2.5075
107
PSS_
-2.5096
108
Garros Solo Lab
-2.5139
109
GXR123
-2.5169
110
XIII
-2.5172
111
ARKANE
-2.5209
112
Eastern Eagles
-2.5246
113
Nietzsche
-2.5272
114
Try hard
-2.5347
115
打枣大队
-2.5379
116
AbangFrabi
-2.5381
117
Banana Kingdom
-2.5388
118
Jade
-2.5458
119
katharsis
-2.5473
120
JiaxinLi
-2.5482
121
Salt-Lake-Agent
-2.5534
122
Mars
-2.5563
123
Husky River Trading
-2.5567
124
Probably Right
-2.5643
125
Liqourena
-2.5769
T2 Forecasting: teams 126–150 of 154
Rank
Team
Score
126
Launchlane Forecast
-2.5799
127
3Dogs Nexus
-2.5813
128
team-5c45a79157100c4d96f055d0e219d78b
-2.5920
129
NSYSU Quant-1
-2.6067
130
Luca
-2.6137
131
lwz888
-2.6141
132
Undefined
-2.6151
133
Tipipali
-2.6181
—
Reference forecaster M0 reference, not ranked
-2.6412
134
QQQQ
-2.6714
135
SoloTraveler
-2.6765
136
HaHaHa
-2.6829
136
paperagent
-2.6829
136
ah_a1g
-2.6829
136
Give me a job
-2.6829
140
Speed Measurements
-2.6856
141
OffTheTape
-2.7369
142
BigWorld
-2.7476
143
SZZTZD
-2.7479
144
K-uant
-2.7985
145
PA-Agent
-2.8354
146
team
-2.9680
—
DSS Capital
— not scored
—
9:00pm
— not scored
—
ZOLDIK
— not scored
—
Urithiru
— not scored
T2 Forecasting: teams 151–154 of 154
Rank
Team
Score
—
iuquant
— not scored
—
Edge-X
— not scored
—
Artifical General Invesment
— not scored
—
Outsiders
— not scored
No team matches that search.
T3 Simulation
T3 Simulation: teams 1–25 of 70
Rank
Team
Score
1
fuzz
231024.2407
2
Neko's Life
230927.0937
3
Proof of Alpha
223443.4710
4
VerityLovityFalsityCruelty
222058.1706
5
IMAnonymous
221629.4451
6
abc123
221088.2954
7
Yan Su
220250.6746
8
iMak AI Lab
219113.6012
9
recently rejected
217315.3433
10
Cornfield Chase
215846.5157
11
Warkop PuteraPrakoso
212137.4968
12
LOOOONG
210603.8783
13
PA-Agent
208075.8518
14
DKYnumber1
207368.8332
15
Paragon
207251.5768
16
cache_miss
206945.0911
17
S2WISH
203394.4709
18
Yongchang
202882.2443
19
HackStreet Boys
202117.0056
20
iluss
201810.1820
21
LastDigitsOfPi
199483.3952
22
Polaris
199431.0656
23
jinfanhahaha
198985.9612
24
Masato
193956.4388
25
Quantumonster
193044.3011
T3 Simulation: teams 26–50 of 70
Rank
Team
Score
26
Quant Science Flow
192022.6929
27
Money Miner
189247.1050
28
Made in Heaven
188021.6305
29
lingsio
187233.1800
30
Autonomous Alpha
185936.6020
31
Lumia
185889.4276
32
PumpkinChicken
185847.8278
33
F1_1
184212.9030
34
Sequoia
181512.8417
35
Verifiable Capital
180949.6635
36
huhudawang
178805.3719
37
Jin & Pei
176009.6963
38
GenerationalUnawareness
175874.1857
39
Probably Right
172355.6323
40
1991
172214.7001
41
Saifuddin
168732.8315
42
EDB AI
168452.0763
43
Chirayu
159905.6425
44
Hello
153910.5377
45
fiftyfifty
152138.2119
46
NolanT
150055.5915
47
DSS Capital
135870.3124
48
YuLiang
133561.8955
49
chilli
121630.3250
50
Auror
110432.6290
T3 Simulation: teams 51–70 of 70
Rank
Team
Score
51
Zeal
28973.5745
52
Tali Labs
18096.1729
53
Apex
16467.6082
54
Eastern Eagles
16411.3761
55
CMCCGDYDDICT
15559.8991
56
tokenmaxxing
15246.9010
57
Banana Kingdom
14180.6623
58
Just-in-Time
13813.8474
59
Shepherd Team
13627.6075
60
Orrery
12998.5100
61
Outsiders
12116.2977
62
RRRRRR
11810.4422
63
3Dogs Nexus
11676.2211
64
thalachira
616.6966
—
TEst
— not scored
—
Hertz
— not scored
—
juniorquant!
— not scored
—
Artifical General Invesment
— not scored
—
oooooo
— not scored
—
Jade
— not scored
No team matches that search.
T4 Explainability
T4 Explainability: teams 1–25 of 69
Rank
Team
Score
1
Saifuddin
0.9643
2
Lucky
0.9312
3
Nietzsche
0.8769
4
Genshin, launch!
0.8621
5
FAWDA
0.8203
6
Yan Su
0.6435
7
huhudawang
0.6389
8
Just-in-Time
0.5935
9
Zeal
0.5876
10
Jin & Pei
0.5725
10
RRRRRR
0.5725
12
LOOOONG
0.5634
13
Proof of Alpha
0.5607
14
MALIU
0.5592
15
littila
0.5590
16
iMak AI Lab
0.5519
17
TradeFlare
0.5438
18
Winner
0.5419
19
AAAA
0.5385
20
polarbear
0.5377
21
HackStreet Boys
0.5311
22
Quiet Signal
0.5285
23
Flowfront
0.5281
24
abc123
0.5231
25
zzsyyy
0.5174
T4 Explainability: teams 26–50 of 69
Rank
Team
Score
26
lingsio
0.5116
27
Sitadel Insecurities
0.5088
28
iluss
0.5002
29
1991
0.4991
30
BaLaBaLa
0.4967
31
Paragon
0.4920
32
stone stone
0.4902
33
EDB AI
0.4887
34
Northline
0.4797
35
OmniSync
0.4751
36
VIRAT18
0.4720
37
Masato
0.4708
38
yohoho
0.4696
39
huahua
0.4623
40
VerityLovityFalsityCruelty
0.4604
41
WEI-SPBU
0.4589
42
Hatenabase
0.4580
43
Bambampurin
0.4570
44
jinfanhahaha
0.4414
45
Pluto
0.4407
46
Quant Science Flow
0.4387
47
Verifiable Capital
0.4323
48
Orrery
0.4289
49
GANisok
0.4214
50
Rocky & Gary
0.4162
T4 Explainability: teams 51–69 of 69
Rank
Team
Score
51
ccczy
0.4029
52
SignalCraft
0.3993
53
Richard Zhu
0.3986
54
Chaewon-Research
0.3978
55
Agentic AI-Driven Human(黄金暴龙)
0.3958
56
Apex
0.3954
57
Agent007
0.3953
58
Faithful
0.3944
59
DeltaNLP
0.3361
60
Agenthon_42
0.3094
61
IMPEL
0.3023
62
jusep
0.2887
63
BIN Han
0.2868
64
3Dogs Nexus
0.2836
65
The Risk Museum
0.2777
66
scu_statistics
0.2513
67
CtheS
0.2497
68
Probably Right
0.1738
69
Teamvitali
-0.0879
No team matches that search.
2026 phases
Development, then Final + Verification.
Development runs through 12 October 2026. The joint Final + Verification phase runs from 13 to 25 October 2026. Registration and Development close on 12 October at 23:59 Anywhere on Earth (AoE, UTC−12). Final uploads close on 25 October at 23:59 AoE (26 October, 11:59 UTC). An upload made by then runs and counts even if its run finishes after the close; organizers verify the top entries after their runs, so verification can end after 25 October. Keep your image pullable until the results are published.
Upcoming
Development
Build and test with the public starter packages. Check the logged-in website for submission opening status.
0%
complete
-
Upcoming Aug 28 - Oct 12 Development
Teams practice and iterate with the public starter packages. Submission opening is announced separately.
-
Upcoming Oct 13 - Oct 25 Final + Verification
One submission per entered track is evaluated on sealed private-test units. Uploads close on 25 October at 23:59 AoE; organizers verify the top entries after their runs. There is no separate verification submission. Run outputs, logs, and score files stay hidden. Results and awards are published in early November.
Awards
Cash awards in all four tracks.
The top three teams in each track win a cash award, decided by the final standings after verification. That is $3,000 per track and $12,000 in total.
-
1st place
$1,500
In each of the four tracks.
-
2nd place
$1,000
In each of the four tracks.
-
3rd place
$500
In each of the four tracks.
Amounts are in US dollars. A team's award is split equally among its registered members. Eligibility, verification, tax and payment requirements are set by the official rules, and a confirmed winner who accepts an award publishes the code needed to reproduce the method.
Scientific outputs
Agenthon is designed to explain failures, not just rank winners.
Cross-track failure map
Gate failures are labeled and aggregated to show where finance agents break down.
Information uplift
T2 investigates whether text and reasoning improve forecasts; this is not a separate scored component.
Speed-realism frontier
T3 measures throughput only after semantic fidelity and stylized facts survive checks.
Faithfulness under embargo
T4 requires evidence-backed predictions that do not cite future or unsupported facts.
Call for papers
The call for papers is closed.
Accepted papers will be presented as posters at the Agenthon workshop at NeurIPS in Atlanta on Saturday 12 December. The program will follow.
