Skip to main content

Model comparison

Hy3 Preview vs Step 3.7 Flash

Data verified

Head-to-head evidence from 16 shared benchmark results across 4 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

53/100
Margin
2.2pts
← winning
50.76/100
1 category wins1 category wins

BenchAlign evidence: Hy3 Preview not scored; Step 3.7 Flash estimated. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Hy3 Preview and Step 3.7 Flash share 16 comparable benchmark results. 2 of 8 categories are comparable. 6 results are unique to Hy3 Preview; 14 to Step 3.7 Flash.

Updated July 16, 2026
Shared results
16
Hy3 Preview only
6
Step 3.7 Flash only
14
Comparable categories
2 / 8

Pick Step 3.7 Flash if you want the stronger benchmark profile. Hy3 Preview only becomes the better choice if coding is the priority or you want the cheaper token bill.

Confidence note. This is a partial-evidence comparison with 16 shared benchmark results across 4 evidence categories; 2 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Step 3.7 Flash is clearly ahead on the provisional aggregate, 57 to 53. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Step 3.7 Flash's sharpest advantage is in agentic, where it averages 66.4 against 54.4. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 54.4% to 59.5%. Hy3 Preview does hit back in coding, so the answer changes if that is the part of the workload you care about most.

Step 3.7 Flash is also the more expensive model on tokens at $0.20 input / $1.15 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Hy3 Preview. That is roughly Infinityx on output cost alone.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for Hy3 Preview and Step 3.7 Flash
CategoryHy3 PreviewΔStep 3.7 Flash
AgenticHy3 Preview54.4Margin 12.0Step 3.7 Flash66.4
CodingHy3 Preview57.8Margin 1.5Step 3.7 Flash56.3
KnowledgeHy3 Preview33.8MarginNo overlapStep 3.7 FlashNot measured
Inst. FollowingHy3 Preview63.1MarginNo overlapStep 3.7 FlashNot measured

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

More
A · Hy3 PreviewB · Step 3.7 Flash
  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 54.4%B 59.5%
    Winner: Step 3.7 FlashΔ 5.1
    Terminal-Bench 2.0: Hy3 Preview scored 54.4%; Step 3.7 Flash scored 59.5%. Step 3.7 Flash wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

MetricHy3 PreviewStep 3.7 FlashComparison
Input / output priceUSD per 1M tokensHy3 Preview$0 input / $0 outputStep 3.7 Flash$0.2 input / $1.15 outputHy3 Preview has the lower combined listed price.
Generation speedtokens per secondHy3 PreviewNot availableStep 3.7 FlashNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenHy3 PreviewNot availableStep 3.7 FlashNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensHy3 Preview256KStep 3.7 Flash256KListed context windows are equal.

Benchmark Deep Dive

AgenticStep 3.7 Flash wins
BenchmarkHy3 PreviewStep 3.7 FlashResult
Terminal-Bench 2.0Source 54.4%59.5%Step 3.7 Flash leads
Gert LabsSource 36.91%51.57%Step 3.7 Flash leads
AA Agentic IndexSource 30.7%21.5%Hy3 Preview leads
GDPval-AASource 35.7%25.9%Hy3 Preview leads
GDPval-AASource 12141017Hy3 Preview leads
BrowseCompSource 75.8%Not comparable
DeepSearchQASource 92.8%Not comparable
ToolathlonSource 49.5%Not comparable
Claw-EvalSource 67.1%Not comparable
HLE w/ toolsSource 47.2%Not comparable
τ²-bench resultsSource 98.5%Not comparable
APEX-Agents-AASource 14.8%Not comparable
CodingHy3 Preview wins
BenchmarkHy3 PreviewStep 3.7 FlashResult
SWE-bench VerifiedSource 74.4%Not comparable
Terminal-Bench 2.0Source 54.4%59.5%Step 3.7 Flash leads
SciCodeSource 41.2%Not comparable
AA-SciCodeSource 47.6%40.0%Hy3 Preview leads
AA Coding IndexSource 58.8%39.6%Hy3 Preview leads
SWE-bench ProSource 56.3%Not comparable
Terminal-Bench HardSource 35.6%Not comparable
Reasoning
BenchmarkHy3 PreviewStep 3.7 FlashResult
AA-LCRSource 66.7%63.7%Hy3 Preview leads
CritPtSource 4.9%2.3%Hy3 Preview leads
Knowledge
BenchmarkHy3 PreviewStep 3.7 FlashResult
Artificial Analysis Intelligence IndexSource 41.2%30.3%Hy3 Preview leads
GPQASource 87.2%Not comparable
GPQA-DSource 87.2%Not comparable
HLESource 25.5%Not comparable
AA-Omniscience AccuracySource 31.5%25.4%Hy3 Preview leads
AA-Omniscience Hallucination RateSource 73.0%84.4%Hy3 Preview leads
AA-GPQA DiamondSource 89.7%80.9%Hy3 Preview leads
AA-HLESource 31.6%19.9%Hy3 Preview leads
AA-Omniscience IndexSource -18.5%-37.5%Hy3 Preview leads
Multimodal
BenchmarkHy3 PreviewStep 3.7 FlashResult
SimpleVQASource 79.2%Not comparable
V*Source 95.3%Not comparable
AA-MMMU-ProSource 75.3%Not comparable
Design Arena WebsiteSource 1218Not comparable
Inst. Following
BenchmarkHy3 PreviewStep 3.7 FlashResult
IFBenchSource 63.1%Not comparable
AA-IFBenchSource 67.3%Not comparable
Frequently Asked Questions (3)

Which is better, Hy3 Preview or Step 3.7 Flash?

Step 3.7 Flash is ahead on BenchLM's provisional leaderboard, 57 to 53. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 54.4% and 59.5%.

Which is better for coding, Hy3 Preview or Step 3.7 Flash?

Hy3 Preview has the edge for coding in this comparison, averaging 57.8 versus 56.3. Inside this category, AA Coding Index is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Hy3 Preview or Step 3.7 Flash?

Step 3.7 Flash has the edge for agentic tasks in this comparison, averaging 66.4 versus 54.4. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.

Related Comparisons

Last updated: July 16, 2026

The AI models change fast. We track them for you.

A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.