Across 8 comparable benchmarks, Claude 3.5 Sonnet has better reported results on 8 and worse on 0.
Strong evidence8 of 8 shared results countedHow we compare
Claude 3.5 Sonnet is no longer available through its official API. API prices for this version are not shown.
Claude3-Opus is no longer available through its official API. API prices for this version are not shown.
Price basis: Listed rates vs listed rates. The price section below shows the applicable periods.
| Fact | Claude 3.5 Sonnet | Claude3-Opus |
|---|---|---|
| Context window | 200K tokens | 200K tokens |
| Input / 1M | — | — |
| Output / 1M | — | — |
| Blended (4:1) | — | — |
| Status | Retired | Retired |
| Released | June 21, 2024 | March 4, 2024 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 4 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
Claude 3.5 Sonnet ahead on 1 of 1 counted
Claude 3.5 Sonnet ahead on 1 of 1 counted
No fixed-scale average — see the rows in Benchmark results
Claude 3.5 Sonnet ahead on 1 of 1 counted
Claude 3.5 Sonnet ahead on 1 of 1 counted
Claude 3.5 Sonnet ahead on 1 of 1 counted
No shared comparable results for Computer & tool use, Research & search, Long context, Professional work & productivity, Cybersecurity.
Published scores, with matching modes where available. Gaps show Claude 3.5 Sonnet relative to Claude3-Opus. Reading the results
| Benchmark | Claude 3.5 Sonnet | Claude3-Opus | Gap |
|---|---|---|---|
| HumanEvalCode Generation & Editing · Accuracy | 92.0Normal · no tools | 84.9Normal · no tools | +7.1 pts#9 vs #37 of 100 |
| SimpleBenchCommonsense · Score (AVG@5) | 27.5Normal · no tools | 23.5Normal · no tools | 1.17×#89 vs #94 of 106 |
| MATHMathematics · Accuracy | 71.1Normal · no tools | 60.1Normal · no tools | +11.0 pts#18 vs #31 of 42 |
| GPQA DiamondScientific Reasoning · Accuracy | 59.4Normal · no tools | 50.4Normal · no tools | +9.0 pts#165 vs #177 of 200 |
| MMMUVisual Understanding · Accuracy | 68.3Reported best | 59.4Reported best | +8.9 pts#36 vs #45 of 62 |
| MMLUKnowledge Exams · Accuracy | 88.3Normal · no tools | 86.8Normal · no tools | +1.5 pts#16 vs #24 of 114 |
| MMLU-ProKnowledge Exams · Accuracy | 76.1Normal · no tools | 68.5Normal · no tools | +7.7 pts#76 vs #98 of 153 |
| ECICapability Indices · ECI score (capability index, higher is better) | 130Reported best | 127Reported best | 1.02×#117 vs #124 of 168 |
Code Generation & Editing · Accuracy
Commonsense · Score (AVG@5)
Mathematics · Accuracy
Scientific Reasoning · Accuracy
Visual Understanding · Accuracy
A dash means no result is available.
| Benchmark | Claude 3.5 Sonnet | Claude3-Opus |
|---|---|---|
| FrontierMath | 1.0 | — |
| FrontierMath - Tier 4 | 0.0 | — |
| GSM8K | — | 95.0 |
| SWE-bench | 18.1 | — |
| DROP | — | 83.1 |
| MMMU-Pro | 51.5 | — |
| VPCT | 33.0 | — |
| HellaSwag | — | 95.4 |
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%
A blended cost needs a published input and output price on both sides, and one of these two does not have them yet.
Claude 3.5 Sonnet: Listed text rates; pricing tier not confirmed.
Claude3-Opus: Listed text rates; pricing tier not confirmed.
| Price / 1M tokens | Claude 3.5 Sonnet | Claude3-Opus | Ratio |
|---|---|---|---|
| Input | — | — | — |
| Output | — | — | — |
| Cached input | — | — | — |
| Cache write | — | — | — |
| Blended | — | — | — |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Lower prices are highlighted. The cache hit rate applies to input tokens only; models without a published cached-input rate use their normal input price for those tokens.Effective dates: Claude 3.5 Sonnet 2026-11-14; Claude3-Opus 2026-11-14.| Attribute | Claude 3.5 Sonnet | Claude3-Opus |
|---|---|---|
| Availability | Retired | Retired |
| Provider | Anthropic | Anthropic |
| Model type | Multimodal model | Multimodal model |
| First release | June 21, 2024 | March 4, 2024 |
| Context window | 200K tokens | 200K tokens |
| Architecture | Undisclosed | Undisclosed |
| Documented capabilities | Multilingual | Multilingual |
| Input modalities | Text, Image | Text, Image |
| Output modalities | Text | Text |
| Weights | Closed / API only | Closed / API only |
| Commercial use | Proprietary | Proprietary |
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 3.5 Sonnet or Claude3-Opus? 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.