STEMBench ↗2.586 / 4
Hy4 Preview · STEMBench120 · mean progress / 4 · closed ↗
Reported model / MODEL EVIDENCE
Hy4 Preview
24 benchmarks · 26 observations · 26 sources
Scores at a glance
24 benchmarks · large dot: Hy4 Preview; small dots: closest models where reported
Vals Index ↗53.21 %
Hy4 Preview · Vals Index · Accuracy ↗
Vals Finance Index ↗57.51 %
Hy4 Preview · Vals Finance Index · Accuracy ↗
Vals Legal Index ↗71.7 %
Hy4 Preview · Vals Legal Index · Accuracy ↗
Legal Research Bench ↗45.19 %
Hy4 Preview · Legal Research Bench · All-Pass ↗
HLAB-Vals-1.1 ↗8.75 %
Hy4 Preview · HLAB-Vals-1.1 · Task Pass Rate ↗
LegalBench ↗83.76 %
Hy4 Preview · LegalBench · Accuracy ↗
Hy4 Preview · Harvey's Legal Agent Benchmark v1 · Accuracy ↗
Vals Finance Agent v2.1 ↗61.62 %
Hy4 Preview · Finance Agent · Accuracy ↗
Hy4 Preview · Excel Modeling Benchmark · Accuracy ↗
1 more reported result
58.08 % · Hy4 Preview · Excel Modeling Benchmark v1 · Accuracy · Vals AI · Excel Modeling Benchmark v1 ↗
Tax Agent Bench ↗32.62 %
Hy4 Preview · Tax Agent Bench · Accuracy ↗
Vals Finance Agent v2 ↗55.06 %
Hy4 Preview · Finance Agent v2 · Accuracy ↗
MedCode ↗43.25 %
Hy4 Preview · MedCode · Accuracy ↗
MedScribe ↗83.6 %
Hy4 Preview · MedScribe · Accuracy ↗
ProofBench ↗75 %
Hy4 Preview · ProofBench · Accuracy ↗
BioMysteryBench ↗69.26 %
Hy4 Preview · BioMysteryBench · Accuracy ↗
Vibe Code Bench ↗77.48 %
Hy4 Preview (OpenHands) · Vibe Code Bench · original task set ↗
Code Migration v1.1 ↗45.36 %
Hy4 Preview · Code Migration · Accuracy ↗
1 more reported result
47.43 % · Hy4 Preview · Code Migration v1 · Accuracy · Vals AI · Code Migration v1 ↗
Hy4 Preview · ProgramBench · Accuracy ↗
Terminal-Bench 4.0 ↗8.08 %
Hy4 Preview (Mini-SWE-agent) · Terminal-Bench 4.0 · Score ↗
IOI v1 & IOI v2 ↗59.33 %
Hy4 Preview · IOI · Accuracy ↗
CyberBench v1.1 ↗65.36 %
Hy4 Preview · CyberBench · Accuracy ↗
SRE-Bench ↗2.29 %
Hy4 Preview · SRE Bench · Accuracy ↗
Public Benefits Bench v1.1 ↗68.61 %
Hy4 Preview · Public Benefits Bench · Accuracy ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
GLM 5.3 Flash →Z.ai
18 of 23 shared benchmark comparisons
MiMo V2.6 Pro →Xiaomi
19 of 23 shared benchmark comparisons
MiMo V2.6 Flash →Xiaomi
20 of 23 shared benchmark comparisons
How similarity is measured
Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.
Sources
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
- STEMBench — Six-Domain Model Evaluation | Ulam ↗
- BioMysteryBench — evaluation methodology and results ↗
- Code Migration v1 — evaluation methodology and results ↗
- Code Migration — evaluation methodology and results ↗
- CyberBench — evaluation methodology and results ↗
- Excel Modeling Benchmark v1 — evaluation methodology and results ↗
- Excel Modeling Benchmark — evaluation methodology and results ↗
- Finance Agent v2 — evaluation methodology and results ↗
- Finance Agent — evaluation methodology and results ↗
- Vals Finance Index — evaluation methodology and results ↗
- HLAB-Vals-1.1 — evaluation methodology and results ↗
- Harvey's Legal Agent Benchmark v1 — evaluation methodology and results ↗
- IOI — evaluation methodology and results ↗
- LegalBench — evaluation methodology and results ↗
- Vals Legal Index — evaluation methodology and results ↗
- Legal Research Bench — evaluation methodology and results ↗
- MedCode — evaluation methodology and results ↗
- MedScribe — evaluation methodology and results ↗
- ProgramBench — evaluation methodology and results ↗
- ProofBench — evaluation methodology and results ↗
- Public Benefits Bench — evaluation methodology and results ↗
- SRE Bench — evaluation methodology and results ↗
- Tax Agent Bench — evaluation methodology and results ↗
- Terminal-Bench 4.0 — evaluation methodology and results ↗
- Vals Index — evaluation methodology and results ↗
- Vibe Code Bench — evaluation methodology and results ↗
Names used in source tables
Hy4 Preview · Hy4 Preview (Mini-SWE-agent) · Hy4 Preview (OpenHands)