Compare/GPT-4o (May '24) vs Claude Sonnet 4.6 (Non-reasoning, High Effort)

GPT-4o (May '24)vsClaude Sonnet 4.6 (Non-reasoning, High Effort)

Side-by-side comparison of pricing, 12 benchmarks, and generation speed.

OpenAI

GPT-4o (May '24)

Input
$5/M
Output
$15/M
Speed
91 tok/s
TTFT
1.24s
Anthropic

Claude Sonnet 4.6 (Non-reasoning, High Effort)

Input
$3/M
Output
$15/M
Speed
43 tok/s
TTFT
1.36s

Winner by Category

Cheaper
Claude Sonnet 4.6 (Non-reasoning, High Effort)
Faster (tok/s)
GPT-4o (May '24)
Lower Latency
GPT-4o (May '24)
Benchmarks (1-1)
Tie

Pricing Comparison

MetricGPT-4o (May '24)Claude Sonnet 4.6 (Non-reasoning, High Effort)
Input ($/M tokens)$5$3
Output ($/M tokens)$15$15
Cost for 1M input + 100K output tokens:
GPT-4o (May '24)$6.50
Claude Sonnet 4.6 (Non-reasoning, High Effort)$4.50

Speed Comparison

Output Speed (tokens/s) — higher is better
GPT-4o (May '24)
91 tok/s
Claude Sonnet 4.6 (Non-reasoning, High Effort)
43 tok/s
Time to First Token (seconds) — lower is better
GPT-4o (May '24)
1.24s
Claude Sonnet 4.6 (Non-reasoning, High Effort)
1.36s

Editorial Analysis

Verdict. GPT-4o (May '24) and Claude Sonnet 4.6 (Non-reasoning, High Effort) split the benchmark comparison evenly at 1–1. The tiebreaker here is price, speed, and what you actually run them on.

Pricing. Both models sit in the premium bracket for output-token pricing. At 1.0× the per-million-token cost, Claude Sonnet 4.6 (Non-reasoning, High Effort) is meaningfully cheaper if your traffic is output-heavy (long completions, document generation, agent loops). Claude Sonnet 4.6 (Non-reasoning, High Effort) makes more sense when output volume is low and absolute reasoning quality justifies the premium.

Strengths. GPT-4o (May '24) is strongest on Coding Index (24.2), Intelligence Index (8.4). Claude Sonnet 4.6 (Non-reasoning, High Effort) leads on Intelligence Index (36.8).

Speed. On throughput, GPT-4o (May '24) generates tokens at 91 tok/s versus 43 tok/s — about 53% faster. On time-to-first-token, GPT-4o (May '24) responds in 1240ms vs 1360ms, which matters most for chat-style UIs.

Provider. OpenAI and Anthropic sell to overlapping but distinct developer audiences: OpenAI tends to ship frontier reasoning models with premium positioning, while Anthropic often prices more aggressively. Your existing vendor relationships, billing, and SLA preferences may matter as much as the raw numbers above.

Workload cost. Workload scenarios (per million requests at 30M input + 15M output tokens): GPT-4o (May '24) costs $375.00 ($4500/year); Claude Sonnet 4.6 (Non-reasoning, High Effort) costs $315.00 ($3780/year). At a smaller 5M-input/2M-output scale (single-developer tool or prototype): GPT-4o (May '24) ≈ $55.00/run, Claude Sonnet 4.6 (Non-reasoning, High Effort) ≈ $45.00/run. At agent/realtime scale (200M input / 100M output per million requests): GPT-4o (May '24) ≈ $2500/run, Claude Sonnet 4.6 (Non-reasoning, High Effort) ≈ $2100/run. Claude Sonnet 4.6 (Non-reasoning, High Effort) becomes more attractive at higher volume — the absolute per-token pricing difference compounds when you ship at scale.

Recommendation. Both models have legitimate use cases — the right answer depends on whether you are optimizing for benchmark ceiling, latency, or unit cost. Start with the cheaper / faster model, evaluate against your specific task, and only switch if the upgrade shows a meaningful lift.

Head-to-head deltas

  • Benchmark wins tie exactly at 1–1. The tiebreaker on raw benchmark parity will be price, speed, or capability coverage.
  • On throughput, GPT-4o (May '24) is 2.11× faster (91 tok/s vs 43 tok/s). For streaming chat or real-time agents this alone often flips the recommendation.
  • Aggregate benchmark score (sum across 12 categories, capped at 100): GPT-4o (May '24) = 33, Claude Sonnet 4.6 (Non-reasoning, High Effort) = 37. Within 15% — effectively equivalent if both meet the threshold your product requires.

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
8.436.8
Coding Index
24.2
Math Index
GPQA Diamond
MMLU-Pro
LiveCodeBench
AIME 2025
MATH-500
Humanity's Last Exam
SciCode
IFBench
TerminalBench
GPT-4o (May '24)1 wins
1 winsClaude Sonnet 4.6 (Non-reasoning, High Effort)

Frequently Asked Questions

Which is cheaper, GPT-4o (May '24) or Claude Sonnet 4.6 (Non-reasoning, High Effort)?

Claude Sonnet 4.6 (Non-reasoning, High Effort) is cheaper overall. Its blended price (3:1 input/output ratio) is $6.00/M tokens vs $7.50/M for GPT-4o (May '24).

Which model performs better on benchmarks?

It's a tie — both models win 1 benchmarks each across 12 evaluated categories. See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

GPT-4o (May '24) generates tokens faster at 91 tok/s vs 43 tok/s. GPT-4o (May '24) also has lower time-to-first-token (1.24s vs 1.36s).

When should I use GPT-4o (May '24) vs Claude Sonnet 4.6 (Non-reasoning, High Effort)?

Choose based on your priorities: Claude Sonnet 4.6 (Non-reasoning, High Effort) for lower cost, both perform similarly on benchmarks, and GPT-4o (May '24) for faster generation. For latency-sensitive apps, check the TTFT comparison above.