ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

Claude Opus 4vsDeepSeek V3.2

DeepSeek V3.2 leads 2 of 2 shared benchmarks. Claude Opus 4 has the larger context window (200K tokens).

Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.

Shared benchmarks
0 – 2
DeepSeek V3.2 leads
Cheaper per token
list price, input + output
Larger context
Claude Opus 4
200K tokens
Provider uptime (30d)
100% · 100%
Anthropic · DeepSeek
Claude Opus 4
Anthropic · released 2025-05-22
textvision
Context
200K
Max out
32K
Input /1M
$15
Output /1M
$75
Cached /1M
$1.5
Scores
6 · 5 core
DeepSeek V3.2
DeepSeek · released 2025-12-01
text
Context
131K
Max out
66K
Input /1M
Output /1M
Cached /1M
Scores
8 · 7 core
Quality

Benchmark matrix

BenchmarkClaude Opus 4DeepSeek V3.2ΔEdge
Reported by both · 2
SWE-bench Verified72.5% 73.1% -0.6 ptDeepSeek V3.2
GPQA Diamond76.3% epoch82.4% -6.1 ptDeepSeek V3.2
Only Claude Opus 4 reports · 4
AIME 2024 (no extended thinking)33.9% not reported
OTIS Mock AIME 2024-2025 (Epoch AI run)64.4% epochnot reported
SWE-bench Verified (Epoch AI run)70.7% epochnot reported
Terminal-bench43.2% not reported
Only DeepSeek V3.2 reports · 5
MMLU-Pronot reported85%
AIME 2025not reported93.1%
AIME 2026not reported94.2% matharena
Humanity's Last Exam (no tools)not reported25.1%
LMArena Elonot reported1424.8
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is Claude Opus 4 minus DeepSeek V3.2 in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing

Side by side

SpecClaude Opus 4DeepSeek V3.2Edge
Context window200K tokens131K tokensClaude Opus 4
Max output32K tokens66K tokensDeepSeek V3.2
Input price / 1M$15
Output price / 1M$75
Cached input / 1M$1.5
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
$90
Modalitiestext · visiontextClaude Opus 4
Released2025-05-222025-12-01
Cited benchmark scores68
Reliability

Provider status

All providers
More matchups

Claude Opus 4 vs …

Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run Claude Opus 4 and DeepSeek V3.2 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.