# Claude Opus 4.6 vs DeepSeek-V3.2-Exp

> Claude Opus 4.6 is the stronger model overall, scoring 58.3 to 44.9 on the Noometry Index. DeepSeek-V3.2-Exp costs 34× less per token, which makes it the better buy when Claude Opus 4.6's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-opus-4-6-vs-deepseek-v3-2-exp
- Last updated: 2026-10-10
- Shared benchmarks: 42

## Summary

- They share 42 benchmarks with published results for both. Claude Opus 4.6 scores higher in 9 categories and DeepSeek-V3.2-Exp in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 4.6 leads 57.7 to 22.0.
- The biggest single-benchmark swing is ARC-AGI-2: 69.2% for Claude Opus 4.6 and 4% for DeepSeek-V3.2-Exp.
- DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 per million input/output tokens, against $5 / $25 for Claude Opus 4.6.
- Claude Opus 4.6 accepts more context: 1M tokens versus 164K.
- DeepSeek-V3.2-Exp has downloadable open weights; the other is API-only.

## Snapshot

| | Claude Opus 4.6 | DeepSeek-V3.2-Exp |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 58.3 | 44.9 |
| Rank | 18 | 68 |
| Context | 1M | 164K |
| Input $/M | $5 | $0.26 |
| Output $/M | $25 | $0.38 |
| Weights | Proprietary | Open |

## Coding

- Claude Opus 4.6: 57.7 (#15)
- DeepSeek-V3.2-Exp: 46.4 (#61)

| Benchmark | Claude Opus 4.6 | DeepSeek-V3.2-Exp |
|---|---|---|
| SWE-bench Verified (bash only) | 75.6% | 70% |
| LMArena WebDev | 1546 | 1362 |
| SWE-bench Multilingual | 72% | 59% |
| WeirdML | 78% | 39.5% |
| LMArena Coding | 1535 | 1454 |
| SWE-bench Verified | 78.7% | — |
| FrontierCode | 26.6% | — |
| Aider Polyglot | — | 74.2% |
| SciCode | — | 38.9% |
| GSO | 41.2% | — |
| ALE-Bench | 996.5 | — |
| AlgoTune | 1.47 | — |

## Agentic & Tool Use

- Claude Opus 4.6: 51.0 (#4)
- DeepSeek-V3.2-Exp: 37.9 (#31)

| Benchmark | Claude Opus 4.6 | DeepSeek-V3.2-Exp |
|---|---|---|
| Terminal-Bench | 79.8% | 39.6% |
| Vending-Bench 2 | 8,018 | 1,034 |
| APEX-Agents | 46.3% | — |
| Berkeley Function Calling Leaderboard | — | 56.7% |
| Remote Labor Index | 4.2% | — |
| TheAgentCompany | — | 42.9% |
| τ²-bench Banking | 27.3% | — |
| Cybench | 93% | — |
| DeepResearch Bench | 55.3% | — |
| GBAEval | 44.1% | — |
| LMArena Search | 1253 | — |
| METR Time Horizons | 78.9% | — |

## Reasoning

- Claude Opus 4.6: 57.7 (#21)
- DeepSeek-V3.2-Exp: 22.0 (#185)

| Benchmark | Claude Opus 4.6 | DeepSeek-V3.2-Exp |
|---|---|---|
| ARC-AGI-2 | 69.2% | 4% |
| Kagi LLM Benchmark | 83.6% | 52.2% |
| NYT Connections (extended) | 92.1% | 36.7% |
| ARC-AGI-1 | 94% | 57% |
| Chess Puzzles | 17% | 14% |
| Thematic Generalization | 80.6% | 65% |
| LMArena Hard Prompts | 1527 | 1433 |
| DTBench | 91.2% | 87.7% |
| LMCA | 55.8% | 29.1% |
| Epoch Capabilities Index | 155.36 | 146.28 |
| SimpleBench | 67.6% | — |
| CritPt | — | 2.9% |
| EnigmaEval | 7.6% | — |
| EBR-Bench | 12.7% | — |
| Mystery Game Puzzles | 25% | — |
| ForecastBench | 60 | — |

## Math

- Claude Opus 4.6: 62.7 (#25)
- DeepSeek-V3.2-Exp: 40.8 (#75)

| Benchmark | Claude Opus 4.6 | DeepSeek-V3.2-Exp |
|---|---|---|
| MathArena Final-Answer Competitions | 78.5% | 57.7% |
| OTIS Mock AIME 2024-2025 | 94.4% | 87.8% |
| ProofBench | 50% | 8% |
| LMArena Math | 1517 | 1435 |
| FrontierMath (Feb 2025 set) | 40.7% | 22.1% |
| FrontierMath Tier 4 (v1) | 22.9% | 2.1% |
| FrontierMath (Tiers 1-3) | 66% | — |
| FrontierMath Tier 4 | 26.8% | — |

## Knowledge

- Claude Opus 4.6: 61.8 (#23)
- DeepSeek-V3.2-Exp: 51.6 (#61)

| Benchmark | Claude Opus 4.6 | DeepSeek-V3.2-Exp |
|---|---|---|
| GPQA Diamond | 90.5% | 83.4% |
| Vectara Hallucination Rate | 12.2% | 5.3% |
| LMArena Expert | 1546 | 1436 |
| Humanity's Last Exam | 34.4% | — |
| SimpleQA Verified | 47% | — |

## Multimodal

- Claude Opus 4.6: 37.0 (#67)
- DeepSeek-V3.2-Exp: —

| Benchmark | Claude Opus 4.6 | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Vision | 1316 | — |
| Furniture Assembly | 28.3% | — |
| LMArena Document | 1507 | — |

## Multilingual

- Claude Opus 4.6: 58.2 (#4)
- DeepSeek-V3.2-Exp: 52.4 (#85)

| Benchmark | Claude Opus 4.6 | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Non-English | 1490 | 1409 |
| LMArena Chinese | 1551 | 1461 |
| LMArena French | 1513 | 1435 |
| LMArena German | 1504 | 1440 |
| LMArena Japanese | 1484 | 1377 |
| LMArena Korean | 1465 | 1372 |
| LMArena Russian | 1497 | 1423 |
| LMArena Spanish | 1512 | 1441 |

## Instruction Following

- Claude Opus 4.6: 79.7 (#3)
- DeepSeek-V3.2-Exp: 74.7 (#89)

| Benchmark | Claude Opus 4.6 | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Instruction Following | 1523 | 1413 |

## Long Context

- Claude Opus 4.6: 48.1 (#13)
- DeepSeek-V3.2-Exp: 47.6 (#15)

| Benchmark | Claude Opus 4.6 | DeepSeek-V3.2-Exp |
|---|---|---|
| CL-bench | 20.7% | 13.2% |
| CL-bench Life | 17% | 9.5% |
| LMArena Longer Query | 1520 | 1428 |
| Fiction.LiveBench | — | 83.3% |

## Writing & Preference

- Claude Opus 4.6: 73.7 (#9)
- DeepSeek-V3.2-Exp: 62.5 (#73)

| Benchmark | Claude Opus 4.6 | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Text | 1504 | 1425 |
| LMArena Creative Writing | 1506 | 1403 |
| EQ-Bench Creative Writing | 1809 | 1515 |
| LMArena Multi-Turn | 1514 | 1427 |
| EQ-Bench 4 | 1223 | — |

## FAQ

### Is Claude Opus 4.6 better than DeepSeek-V3.2-Exp?

Claude Opus 4.6 is the stronger model overall, scoring 58.3 to 44.9 on the Noometry Index. DeepSeek-V3.2-Exp costs 34× less per token, which makes it the better buy when Claude Opus 4.6's lead doesn't matter for your workload.

### Which is cheaper, Claude Opus 4.6 or DeepSeek-V3.2-Exp?

DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; Claude Opus 4.6 lists at $5 and $25.

### Is Claude Opus 4.6 or DeepSeek-V3.2-Exp better for coding?

Claude Opus 4.6 scores higher on coding benchmarks: 57.7 versus 46.4 in the Noometry coding category.

### Which has the bigger context window?

Claude Opus 4.6 does, with 1M tokens against 164K.

### How many benchmarks do Claude Opus 4.6 and DeepSeek-V3.2-Exp share?

42 benchmarks have published results for both models. Claude Opus 4.6 has 68 scored results on Noometry and DeepSeek-V3.2-Exp has 48.