Skip to main content

Model comparison

Claude Haiku 4.5 vs Claude Sonnet 4.5

Data verified

Head-to-head evidence from 5 shared benchmark results across 4 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

56.58/100
Margin
3.0pts
← winning
53.61/100
0 category wins2 category wins

Public leaderboard positions: Claude Haiku 4.5 #77 (Estimated); Claude Sonnet 4.5 #96 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Claude Haiku 4.5 and Claude Sonnet 4.5 share 5 comparable benchmark results. 2 of 8 categories are comparable. 0 results are unique to Claude Haiku 4.5; 7 to Claude Sonnet 4.5.

Updated July 22, 2026
Shared results
5
Claude Haiku 4.5 only
0
Claude Sonnet 4.5 only
7
Comparable categories
2 / 8

Pick Claude Haiku 4.5 if you want the stronger benchmark profile. Claude Sonnet 4.5 only becomes the better choice if mathematics is the priority.

Confidence note. This is a partial-evidence comparison with 5 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

Claude Haiku 4.5 has the cleaner BenchAlign overall profile here, landing at 56.58 versus 53.61. It is a real lead, but still close enough that category-level strengths matter more than the headline number.

Claude Sonnet 4.5 is also the more expensive model on tokens at $3.00 input / $15.00 output per 1M tokens, versus $1.00 input / $5.00 output per 1M tokens for Claude Haiku 4.5. That is roughly 3.0x 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 Claude Haiku 4.5 and Claude Sonnet 4.5
CategoryClaude Haiku 4.5ΔClaude Sonnet 4.5
MathClaude Haiku 4.54.9Margin 6.3Claude Sonnet 4.511.2
CodingClaude Haiku 4.573.3Margin 3.9Claude Sonnet 4.577.2
AgenticClaude Haiku 4.5Not measuredMarginNo overlapClaude Sonnet 4.555.4
ReasoningClaude Haiku 4.5Not measuredMarginNo overlapClaude Sonnet 4.513.6
KnowledgeClaude Haiku 4.5Not measuredMarginNo overlapClaude Sonnet 4.583.4

Decisive benchmark drivers

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

More
A · Claude Haiku 4.5B · Claude Sonnet 4.5
  1. FrontierMath v2 (Tiers 1-3)

    Math
    Source ↗
    A 5.903%B 13.495%
    Winner: Claude Sonnet 4.5Δ 7.6
    FrontierMath v2 (Tiers 1-3): Claude Haiku 4.5 scored 5.903%; Claude Sonnet 4.5 scored 13.495%. Claude Sonnet 4.5 wins this benchmark.
  2. SWE-bench Verified

    Coding
    Source ↗
    A 73.3%B 77.2%
    Winner: Claude Sonnet 4.5Δ 3.9
    SWE-bench Verified: Claude Haiku 4.5 scored 73.3%; Claude Sonnet 4.5 scored 77.2%. Claude Sonnet 4.5 wins this benchmark.
  3. FrontierMath v2 (Tier 4)

    Math
    Source ↗
    A 2.083%B 4.167%
    Winner: Claude Sonnet 4.5Δ 2.1
    FrontierMath v2 (Tier 4): Claude Haiku 4.5 scored 2.083%; Claude Sonnet 4.5 scored 4.167%. Claude Sonnet 4.5 wins this benchmark.

Operational comparison

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

MetricClaude Haiku 4.5Claude Sonnet 4.5Comparison
Input / output priceUSD per 1M tokensClaude Haiku 4.5$1 input / $5 outputClaude Sonnet 4.5$3 input / $15 outputClaude Haiku 4.5 has the lower combined listed price.
Generation speedtokens per secondClaude Haiku 4.5Not availableClaude Sonnet 4.5Not availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenClaude Haiku 4.5Not availableClaude Sonnet 4.5Not availableA complete latency comparison is not available.
Context windowmaximum listed tokensClaude Haiku 4.5200KClaude Sonnet 4.5200KListed context windows are equal.

Benchmark Deep Dive

Agentic
BenchmarkClaude Haiku 4.5Claude Sonnet 4.5Result
JobBenchSource 16.0%27.7%Claude Sonnet 4.5 leads
Terminal-Bench 2.0Source 50%Not comparable
OSWorld-VerifiedSource 61.4%Not comparable
VITA-BenchSource 17.0%Not comparable
Gert LabsSource 48.51%Not comparable
CodingClaude Sonnet 4.5 wins
BenchmarkClaude Haiku 4.5Claude Sonnet 4.5Result
SWE-bench VerifiedSource 73.3%77.2%Claude Sonnet 4.5 leads
Reasoning
BenchmarkClaude Haiku 4.5Claude Sonnet 4.5Result
ARC-AGI-2Source 13.6%Not comparable
Knowledge
BenchmarkClaude Haiku 4.5Claude Sonnet 4.5Result
GPQASource 83.4%Not comparable
MathClaude Sonnet 4.5 wins
BenchmarkClaude Haiku 4.5Claude Sonnet 4.5Result
FrontierMath v2 (Tiers 1-3)Source 5.903%13.495%Claude Sonnet 4.5 leads
FrontierMath v2 (Tier 4)Source 2.083%4.167%Claude Sonnet 4.5 leads
AIME 2025Source 87%Not comparable
Multimodal
BenchmarkClaude Haiku 4.5Claude Sonnet 4.5Result
Design Arena WebsiteSource 11521219Claude Sonnet 4.5 leads
Frequently Asked Questions (3)

Which is better, Claude Haiku 4.5 or Claude Sonnet 4.5?

Claude Haiku 4.5 is ahead on BenchLM's BenchAlign leaderboard, 56.58 to 53.61. The biggest single separator in this matchup is FrontierMath v2 (Tiers 1-3), where the scores are 5.903% and 13.495%.

Which is better for coding, Claude Haiku 4.5 or Claude Sonnet 4.5?

Claude Sonnet 4.5 has the edge for coding in this comparison, averaging 77.2 versus 73.3. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.

Which is better for math, Claude Haiku 4.5 or Claude Sonnet 4.5?

Claude Sonnet 4.5 has the edge for math in this comparison, averaging 11.2 versus 4.9. Inside this category, FrontierMath v2 (Tiers 1-3) is the benchmark that creates the most daylight between them.

Related Comparisons

Last updated: July 22, 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.