Leading AI in
Predictive, Generative, and Agentic AI.
fev-bench
TimeH2O-v1
Zero-shot forecasting on real time series across many domains and frequencies, scored by probabilistic accuracy against a seasonal-naive baseline. 108.8M parameters.
- Win rate
- 89.60%
- Skill score
- 50.80%
- Parameters
- 108.8M
Social Simulation Arena
H2O AI Super Agent
#3 H2O AI Marina
Entrants forecast the next public release of poll, sentiment and attention series. Forecasts are locked before the answer exists and scored against the published number.
- Pairwise win rate
- 68.99
- Questions answered
- 67
- Ahead of Anthropic
- +24.72
FutureX
H2O AI Super Agent
Agents predict real-world events before they resolve, with a new board every week, scored once the answers are known. H2O AI Super Agent has earned 27 FutureX medals (9 first places) across 46 weekly boards entered.
- Latest weekly board
- #3
- FutureX medals
- 27
- First places
- 9
- Boards with a medal
- 20 / 46
ImageJevBench Box Track
H2O-Lightning-4B v1.4.1
Models answer with a bounding box and a confidence. Ranks are among 20 ranked systems, scored on 300 public items (546 sealed items are held back for official runs). A preview track: it does not change ImageJevBench's own ranking.
- Capability
- 84.90
- Median latency
- 0.28s
- Per 1,000 boxes
- $0.096
JevBench
H2O-Lightning-31B
Decision models get a situation and typed questions and must return calibrated answers, scored on intelligence, calibration, speed and cost. A 31B model with open weights.
- Capability score
- 84.34
- Latency incl. network
- 0.28s
- Per 1,000 decisions
- $0.029
fev-bench
TimeH2O-v1
Zero-shot forecasting on real time series across many domains and frequencies, scored by probabilistic accuracy against a seasonal-naive baseline. 108.8M parameters.
- Win rate
- 89.60%
- Skill score
- 50.80%
- Parameters
- 108.8M
- #1
Continuous Intelligenceci-proto - #2
H2O.aiTimeH2O-v1
- #3
Mistral AITimestral-209M
- #4
The Forecasting Companyt0-gamma - #5
Abel AI LabSimons-297M - #6
GoogleTimesFM-3
- #7
Walmart LabsiAmTime-base-forecast
- #8
AWSChronos-2
- #9
The Forecasting Companyt0-beta - #10
DatadogToto-2.0-2.5B
ImageJevBench
H2O-Lightning-31B
Image decision models answer typed questions about a picture and must return calibrated answers, scored on intelligence, calibration, speed and cost. An open-weights model.
- Capability score
- 80.88
- Latency incl. network
- 0.59s
- Per 1,000 decisions
- $0.075
- #1
H2O.aiH2O-Lightning-31B
- #2
Sqwishkettei-fat - #3
Levanto LabsSage 1.4 + images, beta - #4
Nace AIDrex Max
- #5
TypeSafe AIJev 1.13.0 + images, preview - #6
AutoloopsAutoloops – Gemma 4 31B IT
- #7
Mohit GargImajev-4B - #8
Mohit Gargimajev 9B - #9
Sqwishkettei-chub - #10
H2O.aiH2O-Lightning-4B v1.4.1
ImageJevBench Box Track
H2O-Lightning-4B v1.4.1
Models answer with a bounding box and a confidence. Ranks are among 20 ranked systems, scored on 300 public items (546 sealed items are held back for official runs). A preview track: it does not change ImageJevBench's own ranking.
- Capability
- 84.90
- Median latency
- 0.28s
- Per 1,000 boxes
- $0.096
- #1
Z.aiGLM-4.6V
- #2
QwenQwen3-VL-32B Instruct
- #3
H2O.aiH2O-Lightning-4B v1.4.1
- #4
AnthropicClaude Haiku 5.5
- #5
GoogleGemma 4 31B IT
- #6
QwenQwen3-VL-235B-A22B Instruct
- #7
QwenQwen3-VL-30B-A3B Instruct
- #8
TypeSafe AIJev 1.14.0 - #9
QwenQwen3.5-397B-A17B
- #10
QwenQwen3-VL-8B Instruct
JevBench
H2O-Lightning-31B
Decision models get a situation and typed questions and must return calibrated answers, scored on intelligence, calibration, speed and cost. A 31B model with open weights.
- Capability score
- 84.34
- Latency incl. network
- 0.28s
- Per 1,000 decisions
- $0.029
- #1
Torchcast AITorchcast Decision 31B v1.0
- #2
Rain WongMetaDecision-31B-Rain
- #3
MachinePulse AIKarpo-Decision-31B - #4
Base Labsmammoth-1 - #5
PelaAIKnowLine 31B-Gen1
- #6
FastinoFastino GLiDE 1.2 - #7
H2O.aiH2O-Lightning-31B v1.0.7
- #8
AgentScopePiccolo 27B
- #9
PerplexityPerplexity Decider v1.1 27B
- #10
vLLM Semantic RoutervLLM-SR d3-turbo
TALENT
TabH2O v1.1
Two public benchmark suites of tabular classification and regression datasets, comparing tabular foundation models, AutoML, and tuned gradient boosting across 300 datasets on TALENT and 51 datasets on TabArena.
- TALENT rank
- #5 / 13
- TALENT datasets
- 300
- TabArena rank
- #7 / 39
- TabArena datasets
- 51
Tabular predictions, one API call away
- #1
InriaTabICL v2fine-tuned
- #2
LG AI ResearchEXAONE-Tabular
- #3
Prior LabsTabPFN v3.0
- #4
InriaTabICL v2
- #5
H2O.aiTabH2O v1.1
- #6
Prior LabsTabPFN v2.6
- #7
Prior LabsTabPFN v2.5
- #8
AWSAutoGluon
- #9
YandexCatBoosttuned
- #10
AWSAutoGluonquick preset
TabArena
TabH2O v1.1
Two public benchmark suites of tabular classification and regression datasets, comparing tabular foundation models, AutoML, and tuned gradient boosting across 300 datasets on TALENT and 51 datasets on TabArena.
- TALENT rank
- #5 / 13
- TALENT datasets
- 300
- TabArena rank
- #7 / 39
- TabArena datasets
- 51
Tabular predictions, one API call away
FutureX
H2O AI Super Agent
Agents predict real-world events before they resolve, with a new board every week, scored once the answers are known. H2O AI Super Agent has earned 27 FutureX medals (9 first places) across 46 weekly boards entered.
- Latest weekly board
- #3
- FutureX medals
- 27
- First places
- 9
- Boards with a medal
- 20 / 46
- #1
Indiv. AI ResearcherOPS-Agentic-Search_S10.2Kimi K3
- #2
Indiv. AI ResearcherSenseFlowDeepSeek-V4-Pro
- #3
H2O.aiH2O AI Super Agent v1.82sonnet-5.5
- #4
Indiv. AI ResearcherRavenobsidian
- #5
Indiv. AI ResearcherOPS-Agentic-Search_S10.7Kimi K3
- #6
Indiv. AI ResearcherOPS-Agentic-Search_S10.3Kimi K3
- #7
Indiv. AI ResearcherSenseFlowdeepseek-v4-pro-v2
- #8
Indiv. AI ResearcherOPS-Agentic-Search_S10.5Kimi K3
- #9
Indiv. AI ResearcherNevar-B harnessGarnet
- #10
Indiv. AI ResearcherOPS-Agentic-Search_S7Kimi K3
Benchmark data read on .
Social Simulation Arena
H2O AI Super Agent
#3 H2O AI Marina
Entrants forecast the next public release of poll, sentiment and attention series. Forecasts are locked before the answer exists and scored against the published number.