vs
0:4
overall capability wins (1 ties across 5 scored). Claude Sonnet 4 leads overall.
- Coding0:2
- Arena0:1
- Speed0:2
- Specs0:1
Verdict
Claude Sonnet 4 leads coding (widest gap: +14.3 on SWE-Rebench); Claude Sonnet 4 ranks higher on LMArena (+48 Elo); Claude Sonnet 4 allows ~7.8x more max output (64K vs 8K).
- Claude Sonnet 4 leads coding (widest gap: +14.3 on SWE-Rebench)
- Claude Sonnet 4 ranks higher on LMArena (+48 Elo)
- Claude Sonnet 4 allows ~7.8x more max output (64K vs 8K)
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- Claude 3.5 Sonnet
- $6.75
- Claude Sonnet 4
- $6.75
Price per 1M tokens
Full comparison
Organization
Anthropic
Anthropic
Family
Claude Sonnet / Haiku
Claude Sonnet / Haiku
License
Proprietary
Proprietary
Open weights
No
No
Release
Jun 20, 2024
May 22, 2025
Knowledge cutoff
-
-
API / provider
Anthropic
Anthropic
Modalities
text, image → text
text, image → text
Specs
Context window
200K
200K
Max output
8K
64K
Parameters
—
—
Pricing
Input $/1M
$3
$3
Output $/1M
$15
$15
Blended $/1M (3∶1)
$6
$6
Speed
tok/s
70
95
TTFT (s)
0.45
0.35
Reasoning
MMLU-Pro
75.1
—
GPQA Diamond
59.9
—
Humanity's Last Exam
3.7
—
AIME 2025
—
—
MATH-500
69.5
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
—
—
AA-LCR v1.1
—
—
CritPt
—
—
MMMU-Pro
—
—
IFBench
—
—
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
33.6
—
SWE-bench Pro
—
42.7
SWE-bench Multilingual
—
—
LiveCodeBench
48.7
59.4
Terminal-Bench 2.1
—
—
Aider Polyglot
—
56.4
Terminal-Bench 3
—
—
BigCodeBench
46.8
—
SciCode
36.6
—
CursorBench
—
—
SWE-Rebench
30.3
44.6
NL2Repo-Bench
—
—
DeepSWE
—
—
WebDev Arena
—
—
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
1,343
1,391
| Metric | Claude 3.5 Sonnet | Claude Sonnet 4 | Delta |
|---|---|---|---|
| Identity | |||
| Organization | Anthropic | Anthropic | — |
| Family | Claude Sonnet / Haiku | Claude Sonnet / Haiku | — |
| License | Proprietary | Proprietary | — |
| Open weights | No | No | — |
| Release | Jun 20, 2024 | May 22, 2025 | — |
| Knowledge cutoff | - | - | — |
| API / provider | Anthropic | Anthropic | — |
| Modalities | text, image → text | text, image → text | — |
| Specs | |||
| Context window | 200K | 200K | tie |
| Max output | 8K | 64K | B +56K |
| Parameters | — | — | — |
| Pricing | |||
| Input $/1M | $3 | $3 | tie |
| Output $/1M | $15 | $15 | tie |
| Blended $/1M (3∶1) | $6 | $6 | tie |
| Speed | |||
| tok/s | 70 | 95 | B +25 |
| TTFT (s) | 0.45 | 0.35 | B +0.10 |
| Reasoning | |||
| MMLU-Pro | 75.1 | — | — |
| GPQA Diamond | 59.9 | — | — |
| Humanity's Last Exam | 3.7 | — | — |
| AIME 2025 | — | — | — |
| MATH-500 | 69.5 | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | — | — | — |
| AA-LCR v1.1 | — | — | — |
| CritPt | — | — | — |
| MMMU-Pro | — | — | — |
| IFBench | — | — | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | 33.6 | — | — |
| SWE-bench Pro | — | 42.7 | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | 48.7 | 59.4 | B +10.7 pts |
| Terminal-Bench 2.1 | — | — | — |
| Aider Polyglot | — | 56.4 | — |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | 46.8 | — | — |
| SciCode | 36.6 | — | — |
| CursorBench | — | — | — |
| SWE-Rebench | 30.3 | 44.6 | B +14.3 pts |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | — | — |
| WebDev Arena | — | — | — |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | 1,343 | 1,391 | B +48 Elo |
Claude 3.5 Sonnet
- Context
- 200K / 8K out
- Parameters
- —
- Price
- $3 / $15
- Speed
- 70 tok/s · 0.45s TTFT
- Modalities
- text, image → text
- License
- Proprietary
Breakout mid-2024 coding model — Artifacts-era default and still a common baseline in evals.
Claude Sonnet 4
- Context
- 200K / 64K out
- Parameters
- —
- Price
- $3 / $15
- Speed
- 95 tok/s · 0.35s TTFT
- Modalities
- text, image → text
- License
- Proprietary
Prior Sonnet generation still available on Bedrock and Google Cloud after Anthropic API retirement of the primary SKU.
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.