Agentic work
Tool use, computer use, and multi-step task completion
Claude Sonnet 5.5
Claude Sonnet 5.5 leads on the public agentic lane, 66.1 to 61.6, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limited
Claude Opus 4.8 vs Claude Sonnet 5.5
Updated September 28, 2026. Rank says Claude Sonnet 5.5 is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.
Both of these models will change. Get the price, version and retirement notices for the pair, each with its source. Follow model changes
Share or export
Decision reading
Claude Sonnet 5.5 has the higher public score estimate, 80.49 versus 70.39, but the 90% score intervals overlap. Treat that as a lead, not a settled winner. 6 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
- Shared results
- 6
- Claude Opus 4.8 only
- 32
- Claude Sonnet 5.5 only
- 41
- Like-for-like categories
- 2 / 8
Supported: Claude Opus 4.8 · Estimated: Claude Sonnet 5.5How the comparison works
Which one for your work
Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.
Chat turn cost
1K fresh input + 500 output tokens
Claude Sonnet 5.5
Claude Sonnet 5.5 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-ratesCache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Claude Sonnet 5.5
Claude Sonnet 5.5 has the lower estimated token cost for this stated workload. Claude Opus 4.8 has no published cached-input rate, so cached tokens use its listed input rate.
Confidence: rate-fallback
Show secondary and unsupported calls
Repository review cost
50K fresh input + 3K output tokens
Claude Sonnet 5.5
Claude Sonnet 5.5 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-ratesCoding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
Claude Sonnet 5.5 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.
Confidence: limitedLong documents
Prompts that approach the documented context limit
No clear pick
The documented context windows are equal.
Confidence: documented
Which one for a specific job
Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.
The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.
62.2Claude Opus 4.879.6Claude Sonnet 5.5
Directional only · BenchAlign v5.7
Claude Sonnet 5.5 scores higher, but at least one score rests on Estimated evidence or a different benchmark set. Directional only, no winner.
Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.
Same basis rules as the category table below
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
32 Claude Opus 4.8 only
6 Shared
41 Claude Sonnet 5.5 only
2 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Shape of the matched evidence
Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.
Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.
SWE-bench ProCoding
Normalized gap 12.1Claude Opus 4.8:69.2%Claude Sonnet 5.5:81.3%HLE w/o toolsKnowledge
Normalized gap 7.1Claude Opus 4.8:49.8%Claude Sonnet 5.5:56.9%SWE MultilingualCoding
Normalized gap 5.9Claude Opus 4.8:84.4%Claude Sonnet 5.5:90.3%OfficeQA ProMultimodal
Normalized gap 0.6Claude Opus 4.8:66.2%Claude Sonnet 5.5:65.6%
Bars run 0–100 on each benchmark’s normalized display scale
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign v5.7 lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.
Agentic
Like-for-like- Claude Opus 4.8
- 61.6
- Supported · #14/111
- Claude Sonnet 5.5
- 66.1
- Supported · #10/111
- Basis
- BenchAlign v5.7 lane · 12 vs 11 public rows
- Reading
- Claude Sonnet 5.5 leads · intervals overlap
Knowledge
Like-for-like- Claude Opus 4.8
- 70.0
- Supported · #12/160
- Claude Sonnet 5.5
- 80.9
- Supported · #6/160
- Basis
- BenchAlign v5.7 lane · 6 vs 16 public rows
- Reading
- Claude Sonnet 5.5 leads · intervals overlap
Coding
Directional only- Claude Opus 4.8
- 62.2
- Supported · #13/136
- Claude Sonnet 5.5
- 79.6
- Estimated · #3/136
- Basis
- BenchAlign v5.7 lane · 10 vs 9 public rows
- Reading
- Directional only
Reasoning
Directional only- Claude Opus 4.8
- 59.1
- #22/27
- Claude Sonnet 5.5
- 79.2
- #7/27
- Basis
- Provisional lane · 2 vs 0 weighted rows
- Reading
- Directional only
Multimodal
Not comparable- Claude Opus 4.8
- 87.7
- #7/50
- Claude Sonnet 5.5
- 83.3
- Unranked · 7 rankable rows
- Basis
- Provisional lane · 2 vs 1 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Claude Opus 4.8
- Not ranked
- Claude Sonnet 5.5
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Claude Opus 4.8
- 74.0
- #60/124
- Claude Sonnet 5.5
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- Claude Opus 4.8
- 65.2
- #2/7
- Claude Sonnet 5.5
- Not ranked
- Basis
- Provisional lane · 3 vs 0 weighted rows
- Reading
- Not comparable
| Category | Claude Opus 4.8 | Claude Sonnet 5.5 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 61.6Supported · #14/111 | 66.1Supported · #10/111 | Like-for-likeBenchAlign v5.7 lane · 12 vs 11 public rows | Claude Sonnet 5.5 leads · intervals overlap |
| Knowledge | 70.0Supported · #12/160 | 80.9Supported · #6/160 | Like-for-likeBenchAlign v5.7 lane · 6 vs 16 public rows | Claude Sonnet 5.5 leads · intervals overlap |
| Coding | 62.2Supported · #13/136 | 79.6Estimated · #3/136 | Directional onlyBenchAlign v5.7 lane · 10 vs 9 public rows | Directional only |
| Reasoning | 59.1#22/27 | 79.2#7/27 | Directional onlyProvisional lane · 2 vs 0 weighted rows | Directional only |
| Multimodal | 87.7#7/50 | 83.3Unranked · 7 rankable rows | Not comparableProvisional lane · 2 vs 1 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Instruction following | 74.0#60/124 | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | 65.2#2/7 | Not ranked | Not comparableProvisional lane · 3 vs 0 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign v5.7) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
Supported evidence per lane · bars run 0–100Methodology
What each workload costs
Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.
Chat turn
1K fresh input + 500 output tokens
- Claude Opus 4.8
- $0.0175
- Fits in one request
- Claude Sonnet 5.5
- $0.007
- Fits in one request
Claude Sonnet 5.5 has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- Claude Opus 4.8
- $0.325
- Fits in one request
- Claude Sonnet 5.5
- $0.13
- Fits in one request
Claude Sonnet 5.5 has the lower modeled cost
Costs use the listed standard API rates.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Claude Opus 4.8
- $1.35
- Fits in one request
- Cached input priced at the published list-input rate
- Claude Sonnet 5.5
- $0.18
- Fits in one request
Claude Sonnet 5.5 has the lower modeled cost
Claude Opus 4.8 has no published cached-input rate, so cached tokens use its listed input rate.
Cached input falls back to the list input rate only where a cached rate is unpublished
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationClaude Opus 4.8Claude Sonnet 5.5
Context window
Maximum documented context; output-token limits may be lower.
Claude Opus 4.8
Claude Sonnet 5.5
API model ID
Claude Opus 4.8
claude-opus-4-8
Anthropic model overviewClaude Sonnet 5.5
claude-sonnet-5-5
Anthropic Claude Sonnet 5.5 model documentationCached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Claude Opus 4.8
Not published
Claude Sonnet 5.5
$0.2 per 1M cached input tokens
Claude API pricingDocumented inputs
Claude Opus 4.8
text, image
Anthropic model overviewClaude Sonnet 5.5
Documented outputs
Claude Opus 4.8
Claude Sonnet 5.5
Provider availability
Claude Opus 4.8
Generally Available · Claude API
Anthropic model overviewClaude Sonnet 5.5
Generally Available · Claude API, Amazon Bedrock, Google Cloud, Microsoft Foundry, Claude Platform on AWS
Anthropic Claude Sonnet 5.5 model documentationReasoning profile
Claude Opus 4.8
Reasoning
Claude Sonnet 5.5
Reasoning
Weight access
Claude Opus 4.8
Proprietary
Claude Sonnet 5.5
Proprietary
License
Claude Opus 4.8
Proprietary
Claude Sonnet 5.5
Proprietary
Release date
Claude Opus 4.8
2026-05-28
Claude Sonnet 5.5
2026-09-28
If you already use one of these models
- Deployment change
- Both entries list Anthropic as the provider. Confirm endpoint, model ID, limits, and feature support before switching.
- Quality signal
- Claude Sonnet 5.5 has the higher public score estimate, 80.49 versus 70.39, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.325 vs $0.13. Cache-heavy agent loop: $1.35 vs $0.18.
- Context tradeoff
- Both models list 1M.
Run the same representative tasks against both endpoints before changing production traffic.
Questions
Which is better, Claude Opus 4.8 or Claude Sonnet 5.5?
Claude Sonnet 5.5 has the higher public score estimate, 80.49 versus 70.39, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, Claude Opus 4.8 or Claude Sonnet 5.5?
Claude Sonnet 5.5 scores higher for coding on the public lane, 79.6 to 62.2. Claude Sonnet 5.5 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.
Which is better for agentic tasks, Claude Opus 4.8 or Claude Sonnet 5.5?
Claude Sonnet 5.5 leads the public agentic tasks lane, 66.1 to 61.6, with Supported evidence for both models, although the 90% intervals overlap.
Which costs less, Claude Opus 4.8 or Claude Sonnet 5.5?
For the stated presets, chat costs $0.0175 on Claude Opus 4.8 and $0.007 on Claude Sonnet 5.5; repository review costs $0.325 and $0.13; the cache-heavy agent loop costs $1.35 and $0.18. Claude Opus 4.8 has no published cached-input rate, so cached tokens use its listed input rate.
Which has the larger context window, Claude Opus 4.8 or Claude Sonnet 5.5?
Both models list the same context window, 1M.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence79 rows
Agentic
Terminal-Bench 3.0
Claude Opus 4.821.1%SourceClaude Sonnet 5.5—Not directly comparable
Terminal-Bench 2.1
Claude Opus 4.874.6%SourceClaude Sonnet 5.5—Not directly comparable
BrowseComp
Claude Opus 4.884.3%SourceClaude Sonnet 5.5—Not directly comparable
DeepSearchQA
Claude Opus 4.893.1%SourceClaude Sonnet 5.5—Not directly comparable
OSWorld-Verified
Claude Opus 4.883.4%SourceClaude Sonnet 5.5—Not directly comparable
Finance Agent v2
Claude Opus 4.853.9%SourceClaude Sonnet 5.5—Not directly comparable
MCP Atlas
Claude Opus 4.882.2%SourceClaude Sonnet 5.5—Not directly comparable
Toolathlon
Claude Opus 4.859.9%SourceClaude Sonnet 5.5—Not directly comparable
Gert Labs
Claude Opus 4.872.97%SourceClaude Sonnet 5.5—Not directly comparable
ResearchClawBench
Claude Opus 4.821.1%SourceClaude Sonnet 5.5—Not directly comparable
OSWorld 2.0
Claude Opus 4.820.6%SourceClaude Sonnet 5.5—Not directly comparable
Terminal-Bench 2.1 (Vals)
Claude Opus 4.871.9%SourceClaude Sonnet 5.5—Not directly comparable
Terminal-Bench 4.0
Claude Opus 4.8—Claude Sonnet 5.570.60%SourceNot directly comparable
HLE w/ tools
Claude Opus 4.8—Claude Sonnet 5.564.5%SourceNot directly comparable
DRACO
Claude Opus 4.8—Claude Sonnet 5.587.0%SourceNot directly comparable
Terminal-Bench-Science 0.1
Claude Opus 4.8—Claude Sonnet 5.559.9%SourceNot directly comparable
LAB all-pass (Harvey held-out)
Claude Opus 4.8—Claude Sonnet 5.510.0%SourceNot directly comparable
LAB criterion-pass (Harvey held-out)
Claude Opus 4.8—Claude Sonnet 5.593.1%SourceNot directly comparable
Toolathlon Verified Pass@3
Claude Opus 4.8—Claude Sonnet 5.585.2%SourceNot directly comparable
Toolathlon Verified Pass³
Claude Opus 4.8—Claude Sonnet 5.568.5%SourceNot directly comparable
Toolathlon Verified avg. turns
Claude Opus 4.8—Claude Sonnet 5.531.6 turnsSourceNot directly comparable
AutomationBench (Zapier 1.0.6)
Claude Opus 4.8—Claude Sonnet 5.544.7%SourceNot directly comparable
Toolathlon-Verified
Claude Opus 4.8—Claude Sonnet 5.577.8%SourceNot directly comparable
Coding
SWE-bench Verified
Claude Opus 4.888.6%SourceClaude Sonnet 5.5—Not directly comparable
SWE-bench Pro
Claude Opus 4.869.2%SourceClaude Sonnet 5.581.3%SourceClaude Sonnet 5.5 leads this result
SWE Multilingual
Claude Opus 4.884.4%SourceClaude Sonnet 5.590.3%SourceClaude Sonnet 5.5 leads this result
SWE Multimodal
Claude Opus 4.838.4%SourceClaude Sonnet 5.554.3%SourceClaude Sonnet 5.5 leads this result
Terminal-Bench 2.1
Claude Opus 4.874.6%SourceClaude Sonnet 5.5—Not directly comparable
cursorBench31
Claude Opus 4.858.4%SourceClaude Sonnet 5.5—Not directly comparable
cursorBench32
Claude Opus 4.862.3%SourceClaude Sonnet 5.5—Not directly comparable
FrontierCode 1.1 Main
Claude Opus 4.846.5%SourceClaude Sonnet 5.546.2%SourceClaude Opus 4.8 leads this result
LiveCodeBench (Vals)
Claude Opus 4.887.8%SourceClaude Sonnet 5.5—Not directly comparable
SWE-bench (Vals)
Claude Opus 4.888.6%SourceClaude Sonnet 5.5—Not directly comparable
cursorBench40
Claude Opus 4.8—Claude Sonnet 5.555.5%SourceNot directly comparable
DeepSWE
Claude Opus 4.8—Claude Sonnet 5.571.0%SourceNot directly comparable
FrontierCode 1.1 Extended
Claude Opus 4.8—Claude Sonnet 5.559.1%SourceNot directly comparable
ProgramBench
Claude Opus 4.8—Claude Sonnet 5.579.7%SourceNot directly comparable
FrontierSWE v2
Claude Opus 4.8—Claude Sonnet 5.561.9%SourceNot directly comparable
Reasoning
Multimodal
OfficeQA Pro
Claude Opus 4.866.2%SourceClaude Sonnet 5.565.6%SourceClaude Opus 4.8 leads this result
ScreenSpot Pro
Claude Opus 4.887.9%SourceClaude Sonnet 5.5—Not directly comparable
CharXiv
Claude Opus 4.889.9%SourceClaude Sonnet 5.5—Not directly comparable
CharXiv w/o tools
Claude Opus 4.880.5%SourceClaude Sonnet 5.5—Not directly comparable
Chartography (no tools)
Claude Opus 4.8—Claude Sonnet 5.561.6%SourceNot directly comparable
Chartography (tools)
Claude Opus 4.8—Claude Sonnet 5.590.2%SourceNot directly comparable
BenchCAD Vision2Code (no tools)
Claude Opus 4.8—Claude Sonnet 5.50.747SourceNot directly comparable
BenchCAD Vision2Code (tools)
Claude Opus 4.8—Claude Sonnet 5.50.963SourceNot directly comparable
Biomedical image analysis
Claude Opus 4.8—Claude Sonnet 5.572.2%SourceNot directly comparable
OfficeQA
Claude Opus 4.8—Claude Sonnet 5.576.9%SourceNot directly comparable
Knowledge
GPQA
Claude Opus 4.893.6%SourceClaude Sonnet 5.5—Not directly comparable
GPQA-D
Claude Opus 4.893.6%SourceClaude Sonnet 5.5—Not directly comparable
HLE
Claude Opus 4.857.9%SourceClaude Sonnet 5.5—Not directly comparable
HLE w/o tools
Claude Opus 4.849.8%SourceClaude Sonnet 5.556.9%SourceClaude Sonnet 5.5 leads this result
GPQA Diamond (Vals)
Claude Opus 4.892.4%SourceClaude Sonnet 5.5—Not directly comparable
MMLU-Pro (Vals)
Claude Opus 4.889.6%SourceClaude Sonnet 5.5—Not directly comparable
HealthBench (raw)
Claude Opus 4.8—Claude Sonnet 5.569.4%SourceNot directly comparable
HealthBench (length-adjusted)
Claude Opus 4.8—Claude Sonnet 5.565.4%SourceNot directly comparable
HealthBench Professional (raw)
Claude Opus 4.8—Claude Sonnet 5.577.1%SourceNot directly comparable
HealthBench Professional
Claude Opus 4.8—Claude Sonnet 5.569.2%SourceNot directly comparable
BioMysteryBench (human-solvable)
Claude Opus 4.8—Claude Sonnet 5.589.2%SourceNot directly comparable
BioMysteryBench (human-difficult)
Claude Opus 4.8—Claude Sonnet 5.544.7%SourceNot directly comparable
SpatialBench Verified
Claude Opus 4.8—Claude Sonnet 5.572.5%SourceNot directly comparable
SingleCellBench
Claude Opus 4.8—Claude Sonnet 5.559.1%SourceNot directly comparable
Protein Design
Claude Opus 4.8—Claude Sonnet 5.551.0%SourceNot directly comparable
Morphology-to-molecule matching
Claude Opus 4.8—Claude Sonnet 5.525.0%SourceNot directly comparable
Medicinal chemistry
Claude Opus 4.8—Claude Sonnet 5.565.3%SourceNot directly comparable
Protein Design library ranking
Claude Opus 4.8—Claude Sonnet 5.554.8%SourceNot directly comparable
De novo protein-binder design
Claude Opus 4.8—Claude Sonnet 5.582.3%SourceNot directly comparable
Protocols (troubleshooting)
Claude Opus 4.8—Claude Sonnet 5.567.3%SourceNot directly comparable
Protocols (understanding)
Claude Opus 4.8—Claude Sonnet 5.566.6%SourceNot directly comparable
Multilingual
Math
USAMO 2026
Claude Opus 4.896.7%SourceClaude Sonnet 5.5—Not directly comparable
FrontierMath v2 (Tiers 1-3)
Claude Opus 4.847.241%SourceClaude Sonnet 5.5—Not directly comparable
FrontierMath v2 (Tier 4)
Claude Opus 4.831.250%SourceClaude Sonnet 5.5—Not directly comparable
ArXivMath Aug. 2026 (no tools)
Claude Opus 4.8—Claude Sonnet 5.586.8%SourceNot directly comparable
ArXivMath Aug. 2026 (tools)
Claude Opus 4.8—Claude Sonnet 5.595.2%SourceNot directly comparable
79 public results · 6 shared
Watch Claude Opus 4.8 vs Claude Sonnet 5.5
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.
Last updated September 28, 2026