Across 16 comparable benchmarks, Claude Sonnet 4.6 has better reported results on 16 and worse on 0. 2 further shared results are listed below, each with the reason it is not counted.
Moderate evidence16 of 18 shared results countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Claude Sonnet 4.6 | DeepSeek V3.2 |
|---|---|---|
| Context window | 1M tokens | 128K tokens |
| Input / 1M | $3.00 | $0.260 |
| Output / 1M | $15.00 | $0.380 |
| Blended (4:1) | $3.46 | $0.190 |
| Status | Available | Available |
| Released | February 17, 2026 | December 1, 2025 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 10 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
Claude Sonnet 4.6 ahead on 5 of 5 counted
Claude Sonnet 4.6 ahead on 1 of 1 counted
Claude Sonnet 4.6 ahead on 1 of 1 counted
Claude Sonnet 4.6 ahead on 2 of 2 counted
Claude Sonnet 4.6 ahead on 2 of 2 counted
Claude Sonnet 4.6 ahead on 1 of 1 counted
Claude Sonnet 4.6 ahead on 1 of 1 counted
No fixed-scale average — see the rows in Benchmark results
No shared comparable results for Long context, Multimodal, Cybersecurity.
Published scores, with matching modes where available. Gaps show Claude Sonnet 4.6 relative to DeepSeek V3.2. Reading the results
| Benchmark | Claude Sonnet 4.6 | DeepSeek V3.2 | Gap |
|---|---|---|---|
| APEX-SWERepository Engineering · Pass@1 | 34.5Thinking High · tools | 25.6≠Reported best | +8.9 pts#43 vs #56 of 61 |
| SWE-bench VerifiedRepository Engineering · Accuracy | 79.6Thinking · no tools | 73.1≠Thinking · tools | +6.5 pts#20 vs #48 of 108 |
| Terminal Bench 2.0Agentic Development · Accuracy | 59.1Thinking · tools | 46.4Thinking · tools | +12.7 pts#20 vs #38 of 46 |
| Terminal Bench HardAgentic Development · Accuracy | 53.0Max · tools | 35.6≠Thinking · tools | +17.4 pts#9 vs #48 of 157 |
| Vibe Code Bench v1.1Code Generation & Editing · Pass rate (%) | 51.5Max · tools | 5.1≠Thinking High · tools | +46.4 pts#61 vs #106 of 119 |
| ARC-AGI-1Abstract Generalization · Score (%) | 86.5Thinking High · tools | 57.0≠Thinking · no tools | +29.5 pts#45 vs #61 of 86 |
| ARC-AGI-2Abstract Generalization · Score (% solved); cost per task (USD)Not counted · Speed and cost telemetry is reported, not scored | 60.4Thinking High · tools | 4.0≠Thinking · no tools | 15.0×#36 vs #71 of 79 |
| Vending-Bench 2Logical Planning · Money Balance (USD)Not counted · Speed and cost telemetry is reported, not scored | 7204Unspecified Reasoning · tools | 1034Unspecified Reasoning · tools | 6.97×#24 vs #54 of 64 |
| FrontierMath - Tier 4Mathematics · Accuracy | 8.3Thinking · no tools · 16K | 2.1≠Thinking · no tools | +6.2 pts#17 vs #31 of 50 |
| CritPtScientific Reasoning · Score | 3.1Max · no tools | 2.9≠Thinking · no tools | 1.07×#78 vs #81 of 134 |
| GPQA DiamondScientific Reasoning · Accuracy | 89.9Thinking · no tools | 82.4Thinking · no tools | +7.5 pts#52 vs #98 of 200 |
| APEX-AgentsTool Orchestration · Score | 43.0Thinking High · tools | 21.3≠Reported best | 2.02×#44 vs #60 of 61 |
| Pinch BenchTool Orchestration · Accuracy | 88.0Thinking · tools | 84.3Thinking · tools | +3.7 pts#6 vs #19 of 38 |
| BrowseCompFact Finding · Accuracy | 74.7Thinking · tools | 51.4≠Thinking · no tools | +23.3 pts#33 vs #44 of 57 |
| Creative WritingWriting | 1811Normal · no tools | 1515Normal · no tools | 1.20×#23 vs #56 of 121 |
| HLEKnowledge Exams · Accuracy | 49.0Thinking · tools | 25.1≠Thinking · no tools | +23.9 pts#53 vs #93 of 151 |
| ECICapability Indices · ECI score (capability index, higher is better) | 152Reported best | 146Reported best | 1.04×#39 vs #69 of 168 |
| LiveBenchCross-capability Suites · Accuracy | 75.3Thinking High · no tools | 62.2≠Thinking · no tools | +13.1 pts#32 vs #76 of 106 |
Repository Engineering · Pass@1
Repository Engineering · Accuracy
Agentic Development · Accuracy
Agentic Development · Accuracy
Code Generation & Editing · Pass rate (%)
Abstract Generalization · Score (%)
Abstract Generalization · Score (% solved); cost per task (USD)
Not counted · Speed and cost telemetry is reported, not scored
Logical Planning · Money Balance (USD)
Not counted · Speed and cost telemetry is reported, not scored
Mathematics · Accuracy
Scientific Reasoning · Score
Scientific Reasoning · Accuracy
Tool Orchestration · Score
Tool Orchestration · Accuracy
Fact Finding · Accuracy
Writing
A dash means no result is available.
| Benchmark | Claude Sonnet 4.6 | DeepSeek V3.2 |
|---|---|---|
| AIME 2026 | — | 94.2 |
| AIME2025 | — | 93.1 |
| HLE-Verified | — | 36.4 |
| MMLU-Pro | — | 85.0 |
| Riemann-bench | — | 8.0 |
| SciCode | 50.1 | — |
| Code Migration | 39.9 | — |
| CodeForces | — | 2386 |
| DeepSWE | 29.9 | — |
| FrontierCode 1.1 | 40.0 | — |
| LiveCodeBench | — | 83.3 |
| SWE-bench Multilingual | — | 59.0 |
| SWE-Bench Pro - Public | — | 40.9 |
| Terminal-Bench 4.0 | 3.0 | — |
| Claw Bench | — | 79.0 |
| MCP-Atlas | 69.5 | — |
| OSWorld 2.0 | 41.5 | — |
| OSWorld-Verified | 72.5 | — |
| PinchBench v2 | 62.7 | — |
| ExploitBench | 25.8 | — |
| LiveBench 2.0 | 65.1 | — |
| Vals Index | 50.6 | — |
| Context Arena | 82.8 | — |
| GDP.pdf | 15.8 | — |
| GeoBench ACW | 73.0 | — |
| MMMU-Pro | 75.6 | — |
| AA-AnalystAgent | 20.0 | — |
| AutomationBench | 5.3 | — |
| EMB (Excel Modeling) | 60.1 | — |
| Finance Agent v2 | 51.0 | — |
| GDPval-AA | 57.0 | — |
| Harvey Lab-AA | 86.0 | — |
| Harvey's Legal Agent Benchmark | 5.0 | — |
| Legal Research Bench | 38.5 | — |
| Public Benefits Bench v1.1 | 62.5 | — |
| SAGE | 46.6 | — |
| Tax Agent Bench | 25.8 | — |
| τ²-Bench | — | 80.3 |
| τ²-Bench - Telecom | 97.9 | — |
| τ³-Banking | 34.4 | — |
Time-of-day providers use peak-hour rates; fixed-price providers use standard rates. Off-peak discounts remain in the published details.
Drag the handle above the bar for cache hits, the one below it for the input : output mix · 5% steps
Blended price = effective input rate at 90% cache hits × 80% + output rate × 20%
Claude Sonnet 4.6 costs1,715%morethan DeepSeek V3.2 per million tokens — 18.2× the price.Using standard rates at this token mix.
Claude Sonnet 4.6: Standard pricing. Anthropic official pricing (platform.claude.com/docs/en/about-claude/pricing) verified 2026-10-21; full 1M context at standard price
DeepSeek V3.2: Standard pricing. DeepInfra official models API (deepseek-ai/DeepSeek-V3.2) verified 2026-11-13
| Price / 1M tokens | Claude Sonnet 4.6 | DeepSeek V3.2 | Ratio |
|---|---|---|---|
| Input | $3.00 | $0.26 | 11.5× |
| Output | $15.00 | $0.38 | 39.5× |
| Cached input | $0.30 | $0.13 | 2.31× |
| Cache write | — | — | — |
| Blended | $3.46 | $0.19 | 18.2× |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Reported cost per task, paired with the score from the same run. Settings and sources may differ; costs do not affect the verdict.
Claude Sonnet 4.6 relative to DeepSeek V3.2
Summary: 1 of 2 shared benchmarks have matching recorded settings and sources. The average is a geometric mean of 1 positive-cost ratios, one equal weight per benchmark. This benchmark set can include related tests and indexes; it is not a workload cost estimate.
Settings or sources differ or are not recorded; interpret the comparison with care.
Claude Sonnet 4.6
DeepSeek V3.2
Effort, protocol, aggregation and source match; other run conditions may differ.
Claude Sonnet 4.6
DeepSeek V3.2
| Benchmark | Claude Sonnet 4.6 | DeepSeek V3.2 |
|---|---|---|
| AIME 2026 | Not reported | |
| AutomationBench | $0.84 / task Average · Pass Rate: 5.0 Thinking High (With Tools) · zapier-private-heldout 1 more settings$0.77 / task Average · Pass Rate: 3.6 Thinking Medium (With Tools) · zapier-private-heldout | Not reported |
| DeepSWE | Not reported | |
| FrontierCode 1.1 | $2.8979 / task Average · Weighted rubric score: 24.3 Max (With Tools) · frontiercode-1-1-main $2.4401 / task Average · Weighted rubric score: 40.0 Max (With Tools) · frontiercode-1-1-extended $2.2694 / task Average · Weighted rubric score: 39.9 Thinking High (With Tools) · frontiercode-1-1-extended 5 more settings$2.6993 / task Average · Weighted rubric score: 23.5 Thinking High (With Tools) · frontiercode-1-1-main $1.4261 / task Average · Weighted rubric score: 30.4 Thinking Low (With Tools) · frontiercode-1-1-extended $1.6481 / task Average · Weighted rubric score: 15.0 Thinking Low (With Tools) · frontiercode-1-1-main $1.9375 / task Average · Weighted rubric score: 37.5 Thinking Medium (With Tools) · frontiercode-1-1-extended $2.3012 / task Average · Weighted rubric score: 21.1 Thinking Medium (With Tools) · frontiercode-1-1-main | Not reported |
| LiveBench 2.0 | $1.6841 / task Average · Overall score (equal-weight mean of 7 suite scores, 0-100): 65.1 Max (task-specific tools) | Not reported |
| Tax Agent Bench | $1.1935 / task Average · Accuracy (%, rubric × citation quality): 25.8 Unspecified Reasoning (With Tools) | Not reported |
| Vibe Code Bench v1.1 | Not reported |
| Attribute | Claude Sonnet 4.6 | DeepSeek V3.2 |
|---|---|---|
| Availability | Available | Available |
| Provider | Anthropic | DeepSeek-AI |
| Model type | Chat model | Reasoning model |
| First release | February 17, 2026 | December 1, 2025 |
| Context window | 1M tokens | 128K tokens |
| Max output | 64K tokens | 8K tokens |
| Total parameters | — | 671B |
| Active parameters | — | 37B |
| Architecture | Undisclosed | Mixture of Experts (MoE) |
| Documented capabilities | Reasoning, Multilingual | Reasoning, Multilingual |
| Knowledge cutoff | 2026-01 | — |
| Input modalities | Text, Image | Text |
| Output modalities | Text | Text |
| Weights | Closed / API only | Open weights |
| Licence | — | MIT License |
| Commercial use | Proprietary | Commercial use allowed |
These are reported-score comparisons, not controlled head-to-head tests.
What people found comparing these models, or using either one. Comments also appear on the models' own pages.
Used Claude Sonnet 4.6 or DeepSeek V3.2? Sign in with Google to share what worked, what did not, and which API or setup you tested.
No community comments on either model yet.
Alternatives with the most benchmark overlap for this matchup.