Coding work
Code generation, repair, and software-engineering tasks
GPT-5.6 Luna
GPT-5.6 Luna leads on the public coding lane, 64.5 to 63.3, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limited
Claude Opus 4.8 vs GPT-5.6 Luna
Updated September 24, 2026. Rank says Claude Opus 4.8 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 Opus 4.8 has the higher public score estimate, 70.46 versus 65.6, but the 90% score intervals overlap. Treat that as a lead, not a settled winner. 18 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
- Shared results
- 18
- Claude Opus 4.8 only
- 20
- GPT-5.6 Luna only
- 11
- Like-for-like categories
- 4 / 8
Supported: Claude Opus 4.8 and GPT-5.6 LunaHow 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.
Agentic work
Tool use, computer use, and multi-step task completion
Claude Opus 4.8
Claude Opus 4.8 leads on the public agentic lane, 61.3 to 55.3, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limitedLong documents
Prompts that approach the documented context limit
GPT-5.6 Luna
GPT-5.6 Luna has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Chat turn cost
1K fresh input + 500 output tokens
GPT-5.6 Luna
GPT-5.6 Luna 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
GPT-5.6 Luna
GPT-5.6 Luna 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-fallbackRepository review cost
50K fresh input + 3K output tokens
GPT-5.6 Luna
GPT-5.6 Luna has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
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.
63.3Claude Opus 4.864.5GPT-5.6 Luna
Like-for-like · BenchAlign v5.7
GPT-5.6 Luna leads the like-for-like coding row, although the 90% intervals overlap.
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.
20 Claude Opus 4.8 only
18 Shared
11 GPT-5.6 Luna only
1 category rests 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.
FrontierMath v2 (Tiers 1-3)Math
Normalized gap 31.4Claude Opus 4.8:47.241%GPT-5.6 Luna:78.600%FrontierMath v2 (Tier 4)Math
Normalized gap 27.3Claude Opus 4.8:31.250%GPT-5.6 Luna:58.500%OSWorld 2.0Agentic
Normalized gap 25.0Claude Opus 4.8:20.6%GPT-5.6 Luna:45.6%ARC-AGI-2Reasoning
Normalized gap 12.5Claude Opus 4.8:72.1%GPT-5.6 Luna:59.5%SWE-bench ProCoding
Normalized gap 6.5Claude Opus 4.8:69.2%GPT-5.6 Luna:62.7%
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
- 61.3
- Supported · #13/105
- GPT-5.6 Luna
- 55.3
- Supported · #28/105
- Basis
- BenchAlign v5.7 lane · 12 vs 9 public rows
- Reading
- Claude Opus 4.8 leads · intervals overlap
Coding
Like-for-like- Claude Opus 4.8
- 63.3
- Supported · #12/135
- GPT-5.6 Luna
- 64.5
- Supported · #9/135
- Basis
- BenchAlign v5.7 lane · 10 vs 7 public rows
- Reading
- GPT-5.6 Luna leads · intervals overlap
Reasoning
Like-for-like- Claude Opus 4.8
- 59.0
- #15/19
- GPT-5.6 Luna
- 54.7
- #18/19
- Basis
- Provisional lane · 2 vs 2 weighted rows
- Reading
- Claude Opus 4.8 leads
Knowledge
Like-for-like- Claude Opus 4.8
- 70.0
- Supported · #11/158
- GPT-5.6 Luna
- 64.6
- Supported · #22/158
- Basis
- BenchAlign v5.7 lane · 6 vs 6 public rows
- Reading
- Claude Opus 4.8 leads · intervals overlap
Multimodal
Directional only- Claude Opus 4.8
- 87.7
- #4/50
- GPT-5.6 Luna
- 67.1
- #22/50
- Basis
- Provisional lane · 2 vs 1 weighted rows
- Reading
- Directional only
Multilingual
Not comparable- Claude Opus 4.8
- Not ranked
- GPT-5.6 Luna
- 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
- GPT-5.6 Luna
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- Claude Opus 4.8
- 65.2
- #2/7
- GPT-5.6 Luna
- 94.2
- Unranked · 3 rankable rows
- Basis
- Provisional lane · 3 vs 2 weighted rows
- Reading
- Not comparable
| Category | Claude Opus 4.8 | GPT-5.6 Luna | Basis | Reading |
|---|---|---|---|---|
| Agentic | 61.3Supported · #13/105 | 55.3Supported · #28/105 | Like-for-likeBenchAlign v5.7 lane · 12 vs 9 public rows | Claude Opus 4.8 leads · intervals overlap |
| Coding | 63.3Supported · #12/135 | 64.5Supported · #9/135 | Like-for-likeBenchAlign v5.7 lane · 10 vs 7 public rows | GPT-5.6 Luna leads · intervals overlap |
| Reasoning | 59.0#15/19 | 54.7#18/19 | Like-for-likeProvisional lane · 2 vs 2 weighted rows | Claude Opus 4.8 leads |
| Knowledge | 70.0Supported · #11/158 | 64.6Supported · #22/158 | Like-for-likeBenchAlign v5.7 lane · 6 vs 6 public rows | Claude Opus 4.8 leads · intervals overlap |
| Multimodal | 87.7#4/50 | 67.1#22/50 | Directional onlyProvisional lane · 2 vs 1 weighted rows | Directional only |
| 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 | 94.2Unranked · 3 rankable rows | Not comparableProvisional lane · 3 vs 2 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
- GPT-5.6 Luna
- $0.0008
- Fits in one request
GPT-5.6 Luna 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
- GPT-5.6 Luna
- $0.0136
- Fits in one request
GPT-5.6 Luna 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
- GPT-5.6 Luna
- $0.02
- Fits in one request
GPT-5.6 Luna 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 4.8GPT-5.6 Luna
Context window
Maximum documented context; output-token limits may be lower.
Claude Opus 4.8
GPT-5.6 Luna
1.05M
OpenAI model catalogAPI model ID
Claude Opus 4.8
claude-opus-4-8
Anthropic model overviewGPT-5.6 Luna
gpt-5.6-luna
OpenAI model catalogCached-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
GPT-5.6 Luna
$0.02 per 1M cached input tokens
OpenAI pricingDocumented inputs
Claude Opus 4.8
text, image
Anthropic model overviewGPT-5.6 Luna
text, image
OpenAI model catalogDocumented outputs
Claude Opus 4.8
GPT-5.6 Luna
Provider availability
Claude Opus 4.8
Generally Available · Claude API
Anthropic model overviewGPT-5.6 Luna
Generally Available · OpenAI Responses API
OpenAI model catalogReasoning profile
Claude Opus 4.8
Reasoning
GPT-5.6 Luna
Reasoning
Weight access
Claude Opus 4.8
Proprietary
GPT-5.6 Luna
Proprietary
License
Claude Opus 4.8
Proprietary
GPT-5.6 Luna
Proprietary
Release date
Claude Opus 4.8
2026-05-28
GPT-5.6 Luna
2026-07-09
If you already use one of these models
- Deployment change
- The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
- Quality signal
- Claude Opus 4.8 has the higher public score estimate, 70.46 versus 65.6, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.325 vs $0.0136. Cache-heavy agent loop: $1.35 vs $0.02.
- Context tradeoff
- GPT-5.6 Luna has the larger documented window (1.05M).
Run the same representative tasks against both endpoints before changing production traffic.
Questions
Which is better, Claude Opus 4.8 or GPT-5.6 Luna?
Claude Opus 4.8 has the higher public score estimate, 70.46 versus 65.6, 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 GPT-5.6 Luna?
GPT-5.6 Luna leads the public coding lane, 64.5 to 63.3, with Supported evidence for both models, although the 90% intervals overlap.
Which is better for agentic tasks, Claude Opus 4.8 or GPT-5.6 Luna?
Claude Opus 4.8 leads the public agentic tasks lane, 61.3 to 55.3, with Supported evidence for both models, although the 90% intervals overlap.
Which costs less, Claude Opus 4.8 or GPT-5.6 Luna?
For the stated presets, chat costs $0.0175 on Claude Opus 4.8 and $0.0008 on GPT-5.6 Luna; repository review costs $0.325 and $0.0136; the cache-heavy agent loop costs $1.35 and $0.02. 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 GPT-5.6 Luna?
GPT-5.6 Luna has the larger documented context window: 1.05M, compared with 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 evidence49 rows
Agentic
Terminal-Bench 3.0
Shared sourceClaude Opus 4.821.1%GPT-5.6 Luna14.3%Claude Opus 4.8 leads this result
Terminal-Bench 2.1
Claude Opus 4.874.6%SourceGPT-5.6 Luna84.7%SourceGPT-5.6 Luna leads this result
BrowseComp
Claude Opus 4.884.3%SourceGPT-5.6 Luna83.3%SourceClaude Opus 4.8 leads this result
DeepSearchQA
Claude Opus 4.893.1%SourceGPT-5.6 Luna—Not directly comparable
OSWorld-Verified
Claude Opus 4.883.4%SourceGPT-5.6 Luna—Not directly comparable
Finance Agent v2
Claude Opus 4.853.9%SourceGPT-5.6 Luna—Not directly comparable
MCP Atlas
Claude Opus 4.882.2%SourceGPT-5.6 Luna—Not directly comparable
Toolathlon
Claude Opus 4.859.9%SourceGPT-5.6 Luna53.4%SourceClaude Opus 4.8 leads this result
Gert Labs
Claude Opus 4.872.97%SourceGPT-5.6 Luna—Not directly comparable
ResearchClawBench
Claude Opus 4.821.1%SourceGPT-5.6 Luna—Not directly comparable
OSWorld 2.0
Claude Opus 4.820.6%SourceGPT-5.6 Luna45.6%SourceGPT-5.6 Luna leads this result
Terminal-Bench 2.1 (Vals)
Claude Opus 4.871.9%SourceGPT-5.6 Luna79.0%SourceGPT-5.6 Luna leads this result
CyberGym
Claude Opus 4.8—GPT-5.6 Luna77.9%SourceNot directly comparable
ExploitGym
Claude Opus 4.8—GPT-5.6 Luna12.4%SourceNot directly comparable
ApprenticeBench
Claude Opus 4.8—GPT-5.6 Luna7%SourceNot directly comparable
Coding
SWE-bench Verified
Claude Opus 4.888.6%SourceGPT-5.6 Luna—Not directly comparable
SWE-bench Pro
Claude Opus 4.869.2%SourceGPT-5.6 Luna62.7%SourceClaude Opus 4.8 leads this result
SWE Multilingual
Claude Opus 4.884.4%SourceGPT-5.6 Luna—Not directly comparable
SWE Multimodal
Claude Opus 4.838.4%SourceGPT-5.6 Luna—Not directly comparable
Terminal-Bench 2.1
Claude Opus 4.874.6%SourceGPT-5.6 Luna84.7%SourceGPT-5.6 Luna leads this result
cursorBench31
Claude Opus 4.858.4%SourceGPT-5.6 Luna—Not directly comparable
cursorBench32
Shared sourceClaude Opus 4.862.3%GPT-5.6 Luna61.1%Claude Opus 4.8 leads this result
FrontierCode 1.1 Main
Claude Opus 4.846.5%SourceGPT-5.6 Luna—Not directly comparable
LiveCodeBench (Vals)
Claude Opus 4.887.8%SourceGPT-5.6 Luna—Not directly comparable
SWE-bench (Vals)
Claude Opus 4.888.6%SourceGPT-5.6 Luna93.0%SourceGPT-5.6 Luna leads this result
DeepSWE
Claude Opus 4.8—GPT-5.6 Luna67.2%SourceNot directly comparable
FrontierCode 1.1 Extended
Claude Opus 4.8—GPT-5.6 Luna55.1%SourceNot directly comparable
VulcanBench v3
Claude Opus 4.8—GPT-5.6 Luna85.5%SourceNot directly comparable
Reasoning
Multimodal
OfficeQA Pro
Claude Opus 4.866.2%SourceGPT-5.6 Luna—Not directly comparable
ScreenSpot Pro
Claude Opus 4.887.9%SourceGPT-5.6 Luna—Not directly comparable
CharXiv
Claude Opus 4.889.9%SourceGPT-5.6 Luna—Not directly comparable
CharXiv w/o tools
Claude Opus 4.880.5%SourceGPT-5.6 Luna—Not directly comparable
MMMU-Pro
Claude Opus 4.8—GPT-5.6 Luna78.4%SourceNot directly comparable
MMMU-Pro w/ Python
Claude Opus 4.8—GPT-5.6 Luna79.5%SourceNot directly comparable
Knowledge
GPQA
Claude Opus 4.893.6%SourceGPT-5.6 Luna92.3%SourceClaude Opus 4.8 leads this result
GPQA-D
Claude Opus 4.893.6%SourceGPT-5.6 Luna92.3%SourceClaude Opus 4.8 leads this result
HLE
Claude Opus 4.857.9%SourceGPT-5.6 Luna—Not directly comparable
HLE w/o tools
Claude Opus 4.849.8%SourceGPT-5.6 Luna—Not directly comparable
GPQA Diamond (Vals)
Claude Opus 4.892.4%SourceGPT-5.6 Luna91.7%SourceClaude Opus 4.8 leads this result
MMLU-Pro (Vals)
Claude Opus 4.889.6%SourceGPT-5.6 Luna86.0%SourceClaude Opus 4.8 leads this result
HealthBench Professional
Claude Opus 4.8—GPT-5.6 Luna55.7%SourceNot directly comparable
HealthBench Hard
Claude Opus 4.8—GPT-5.6 Luna32.0%SourceNot directly comparable
Multilingual
INCLUDE
Claude Opus 4.887.6%SourceGPT-5.6 Luna—Not directly comparable
Math
USAMO 2026
Claude Opus 4.896.7%SourceGPT-5.6 Luna—Not directly comparable
FrontierMath v2 (Tiers 1-3)
Claude Opus 4.847.241%SourceGPT-5.6 Luna78.600%SourceGPT-5.6 Luna leads this result
FrontierMath v2 (Tier 4)
Claude Opus 4.831.250%SourceGPT-5.6 Luna58.500%SourceGPT-5.6 Luna leads this result
FrontierMath (legacy)
Claude Opus 4.8—GPT-5.6 Luna78.6%SourceNot directly comparable
49 public results · 18 shared
Watch Claude Opus 4.8 vs GPT-5.6 Luna
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 24, 2026