Skip to main content
BenchLM
Advertise

Model comparison

Claude Sonnet 4.6 vs o1

Data verified

Head-to-head evidence from 13 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

65.07/100
Margin
17.0pts
← winning
o1
OpenAI
48.1/100
1 category wins1 category wins

Public leaderboard positions: Claude Sonnet 4.6 #32 (Supported); o1 #129 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Claude Sonnet 4.6 and o1 share 13 comparable benchmark results. 2 of 8 categories are comparable. 20 results are unique to Claude Sonnet 4.6; 3 to o1.

Updated July 24, 2026
Shared results
13
Claude Sonnet 4.6 only
20
o1 only
3
Comparable categories
2 / 8

Pick Claude Sonnet 4.6 if you want the stronger benchmark profile. o1 only becomes the better choice if knowledge is the priority or you want the stronger reasoning-first profile.

Confidence note. This is a partial-evidence comparison with 13 shared benchmark results across 6 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 Sonnet 4.6 is clearly ahead on the BenchAlign aggregate, 65.07 to 48.1. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Claude Sonnet 4.6's sharpest advantage is in mathematics, where it averages 26.4 against 9.3. The single biggest benchmark swing on the page is FrontierMath v2 (Tiers 1-3), 32.400% to 9.310%. o1 does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.

o1 is also the more expensive model on tokens at $15.00 input / $60.00 output per 1M tokens, versus $3.00 input / $15.00 output per 1M tokens for Claude Sonnet 4.6. That is roughly 4.0x on output cost alone. o1 is the reasoning model in the pair, while Claude Sonnet 4.6 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use.

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 Sonnet 4.6 and o1
CategoryClaude Sonnet 4.6Δo1
MathClaude Sonnet 4.626.4Margin← 17.1o19.3
KnowledgeClaude Sonnet 4.666.0Margin→ 9.7o175.7
AgenticClaude Sonnet 4.665.2MarginNo overlapo1Not measured
CodingClaude Sonnet 4.669.1MarginNo overlapo1Not measured
MultimodalClaude Sonnet 4.677.4MarginNo overlapo1Not measured
Inst. FollowingClaude Sonnet 4.6Not measuredMarginNo overlapo192.2

Decisive benchmark drivers

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

More
A · Claude Sonnet 4.6B · o1

  1. FrontierMath v2 (Tiers 1-3)

    Math
    Source ↗
    A 32.400%B 9.310%
    Winner: Claude Sonnet 4.6Δ 23.1
    FrontierMath v2 (Tiers 1-3): Claude Sonnet 4.6 scored 32.400%; o1 scored 9.310%. Claude Sonnet 4.6 wins this benchmark.

  2. GPQA

    Knowledge
    Source ↗
    A 89.9%B 75.7%
    Winner: Claude Sonnet 4.6Δ 14.2
    GPQA: Claude Sonnet 4.6 scored 89.9%; o1 scored 75.7%. Claude Sonnet 4.6 wins this benchmark.

Operational comparison

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

MetricClaude Sonnet 4.6o1Comparison
Input / output priceUSD per 1M tokensClaude Sonnet 4.6$3 input / $15 outputo1$15 input / $60 outputClaude Sonnet 4.6 has the lower combined listed price.
Generation speedtokens per secondClaude Sonnet 4.644 tok/so198 tok/so1 has the higher measured throughput.
First-answer latencyseconds to first tokenClaude Sonnet 4.61.48 so132.29 sClaude Sonnet 4.6 reaches the first token sooner.
Context windowmaximum listed tokensClaude Sonnet 4.6200Ko1200KListed context windows are equal.

Benchmark Deep Dive

Agentic
BenchmarkClaude Sonnet 4.6o1Result
Terminal-Bench 2.0Source 59.1%—Not comparable
OSWorld-VerifiedSource 72.1%—Not comparable
Claw-EvalSource 67.8%—Not comparable
CyberGymSource 65.2%—Not comparable
τ²-bench resultsSource 79.5%62.6%Claude Sonnet 4.6 leads
Gert LabsSource 62.92%—Not comparable
OSWorld 2.0Source 8.3%—Not comparable
JobBenchSource 36.9%—Not comparable
Coding
BenchmarkClaude Sonnet 4.6o1Result
SWE-bench VerifiedSource 79.6%—Not comparable
SWE-RebenchSource 60.7%—Not comparable
React Native EvalsSource 80.6%—Not comparable
Vibe Code BenchSource 51.48%—Not comparable
cursorBench31Source 48.8%—Not comparable
AA-SciCodeSource 46.9%35.8%Claude Sonnet 4.6 leads
FrontierCode 1.1 MainSource 24.3%—Not comparable
AA Coding IndexSource —39.7%Not comparable
Reasoning
BenchmarkClaude Sonnet 4.6o1Result
AA-LCRSource 57.7%59.3%o1 leads
CritPtSource 0.9%0.3%Claude Sonnet 4.6 leads
Knowledgeo1 wins
BenchmarkClaude Sonnet 4.6o1Result
GPQASource 89.9%75.7%Claude Sonnet 4.6 leads
SuperGPQASource 95%—Not comparable
MMLU-ProSource 79.2%—Not comparable
HLESource 49%—Not comparable
Artificial Analysis Intelligence IndexSource 35.9%23.4%Claude Sonnet 4.6 leads
AA-GPQA DiamondSource 79.9%74.7%Claude Sonnet 4.6 leads
AA-HLESource 13.2%7.7%Claude Sonnet 4.6 leads
AA-Omniscience IndexSource -2.9%-10.5%Claude Sonnet 4.6 leads
AA-Omniscience AccuracySource 38.0%34.7%Claude Sonnet 4.6 leads
AA-Omniscience Hallucination RateSource 65.9%69.3%Claude Sonnet 4.6 leads
MMLUSource —91.8%Not comparable
MathClaude Sonnet 4.6 wins
BenchmarkClaude Sonnet 4.6o1Result
FrontierMath v2 (Tiers 1-3)Source 32.400%9.310%Claude Sonnet 4.6 leads
FrontierMath v2 (Tier 4)Source 8.300%—Not comparable
Multimodal
BenchmarkClaude Sonnet 4.6o1Result
CharXivSource 77.4%—Not comparable
AA-MMMU-ProSource 70.6%—Not comparable
Design Arena WebsiteSource 1314—Not comparable
Inst. Following
BenchmarkClaude Sonnet 4.6o1Result
AA-IFBenchSource 41.2%70.3%o1 leads
IFEvalSource —92.2%Not comparable
Frequently Asked Questions (3)

Which is better, Claude Sonnet 4.6 or o1?

Claude Sonnet 4.6 is ahead on BenchLM's BenchAlign leaderboard, 65.07 to 48.1. The biggest single separator in this matchup is FrontierMath v2 (Tiers 1-3), where the scores are 32.400% and 9.310%.

Which is better for knowledge tasks, Claude Sonnet 4.6 or o1?

o1 has the edge for knowledge tasks in this comparison, averaging 75.7 versus 66. Inside this category, GPQA is the benchmark that creates the most daylight between them.

Which is better for math, Claude Sonnet 4.6 or o1?

Claude Sonnet 4.6 has the edge for math in this comparison, averaging 26.4 versus 9.3. Inside this category, FrontierMath v2 (Tiers 1-3) is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

Anthropic Compare Pricing Methodology Find Your Best LLM Overall Rankings
Last updated: July 24, 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.