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 50.3, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limited
DeepSeek V4 Pro 0813 vs GPT-5.6 Luna
Updated September 27, 2026. Rank says GPT-5.6 Luna 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
GPT-5.6 Luna has the higher public score estimate, 65.6 versus 63.48, but the 90% score intervals overlap. Treat that as a lead, not a settled winner. 14 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
- Shared results
- 14
- DeepSeek V4 Pro 0813 only
- 28
- GPT-5.6 Luna only
- 15
- Like-for-like categories
- 2 / 8
Estimated: DeepSeek V4 Pro 0813 · Supported: 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.
Long documents
Prompts that approach the documented context limit
GPT-5.6 Luna
GPT-5.6 Luna has the larger documented context window.
Confidence: documentedChat 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-rates
Show secondary and unsupported calls
Cache-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. Costs use the listed standard API rates.
Confidence: listed-ratesRepository 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-ratesAgentic work
Tool use, computer use, and multi-step task completion
No clear pick
The like-for-like agentic result is a practical tie on the public lane (within 0.5 points).
Confidence: limited
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.
50.3DeepSeek V4 Pro 081364.5GPT-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.
28 DeepSeek V4 Pro 0813 only
14 Shared
15 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.
SWE-bench ProCoding
Normalized gap 7.3DeepSeek V4 Pro 0813:55.4%GPT-5.6 Luna:62.7%GPQAKnowledge
Normalized gap 2.2DeepSeek V4 Pro 0813:90.1%GPT-5.6 Luna:92.3%ARC-AGI-2Reasoning
Normalized gap 1.8DeepSeek V4 Pro 0813:61.3%GPT-5.6 Luna:59.5%MMLU-Pro (Vals)Knowledge
Normalized gap 1.0DeepSeek V4 Pro 0813:87.0%GPT-5.6 Luna:86.0%BrowseCompAgentic
Normalized gap 0.1DeepSeek V4 Pro 0813:83.4%GPT-5.6 Luna:83.3%
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- DeepSeek V4 Pro 0813
- 55.0
- Supported · #29/105
- GPT-5.6 Luna
- 55.3
- Supported · #28/105
- Basis
- BenchAlign v5.7 lane · 11 vs 9 public rows
- Reading
- Practical tie
Coding
Like-for-like- DeepSeek V4 Pro 0813
- 50.3
- Supported · #39/135
- GPT-5.6 Luna
- 64.5
- Supported · #9/135
- Basis
- BenchAlign v5.7 lane · 15 vs 7 public rows
- Reading
- GPT-5.6 Luna leads · intervals overlap
Knowledge
Directional only- DeepSeek V4 Pro 0813
- 63.4
- Estimated · #29/158
- GPT-5.6 Luna
- 64.6
- Supported · #22/158
- Basis
- BenchAlign v5.7 lane · 8 vs 6 public rows
- Reading
- Directional only
Reasoning
Not comparable- DeepSeek V4 Pro 0813
- 56.9
- Unranked · 4 rankable rows
- GPT-5.6 Luna
- 54.7
- #18/19
- Basis
- Provisional lane · 1 vs 2 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- DeepSeek V4 Pro 0813
- Not ranked
- GPT-5.6 Luna
- 67.1
- #22/50
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- DeepSeek V4 Pro 0813
- Not ranked
- GPT-5.6 Luna
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- DeepSeek V4 Pro 0813
- Not ranked
- GPT-5.6 Luna
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- DeepSeek V4 Pro 0813
- 80.2
- Unranked · 4 rankable rows
- GPT-5.6 Luna
- 94.2
- Unranked · 3 rankable rows
- Basis
- Provisional lane · 1 vs 2 weighted rows
- Reading
- Not comparable
| Category | DeepSeek V4 Pro 0813 | GPT-5.6 Luna | Basis | Reading |
|---|---|---|---|---|
| Agentic | 55.0Supported · #29/105 | 55.3Supported · #28/105 | Like-for-likeBenchAlign v5.7 lane · 11 vs 9 public rows | Practical tie |
| Coding | 50.3Supported · #39/135 | 64.5Supported · #9/135 | Like-for-likeBenchAlign v5.7 lane · 15 vs 7 public rows | GPT-5.6 Luna leads · intervals overlap |
| Knowledge | 63.4Estimated · #29/158 | 64.6Supported · #22/158 | Directional onlyBenchAlign v5.7 lane · 8 vs 6 public rows | Directional only |
| Reasoning | 56.9Unranked · 4 rankable rows | 54.7#18/19 | Not comparableProvisional lane · 1 vs 2 weighted rows | Not comparable |
| Multimodal | Not ranked | 67.1#22/50 | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Instruction following | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | 80.2Unranked · 4 rankable rows | 94.2Unranked · 3 rankable rows | Not comparableProvisional lane · 1 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
- DeepSeek V4 Pro 0813
- $0.0033
- 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
- DeepSeek V4 Pro 0813
- $0.07788
- 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
- DeepSeek V4 Pro 0813
- $0.0748
- Fits in one request
- GPT-5.6 Luna
- $0.02
- Fits in one request
GPT-5.6 Luna has the lower modeled cost
Costs use the listed standard API rates.
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.
SpecificationDeepSeek V4 Pro 0813GPT-5.6 Luna
Context window
Maximum documented context; output-token limits may be lower.
DeepSeek V4 Pro 0813
GPT-5.6 Luna
1.05M
OpenAI model catalogAPI model ID
DeepSeek V4 Pro 0813
deepseek-v4-pro
DeepSeek models and pricingGPT-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.
DeepSeek V4 Pro 0813
$0.044 per 1M cached input tokens
DeepSeek: Models & PricingGPT-5.6 Luna
$0.02 per 1M cached input tokens
OpenAI pricingDocumented inputs
DeepSeek V4 Pro 0813
GPT-5.6 Luna
text, image
OpenAI model catalogDocumented outputs
DeepSeek V4 Pro 0813
GPT-5.6 Luna
Provider availability
DeepSeek V4 Pro 0813
Generally Available · DeepSeek API
DeepSeek models and pricing, V4 Pro continuation footnoteGPT-5.6 Luna
Generally Available · OpenAI Responses API
OpenAI model catalogReasoning profile
DeepSeek V4 Pro 0813
Reasoning
GPT-5.6 Luna
Reasoning
Weight access
DeepSeek V4 Pro 0813
Open Weight
GPT-5.6 Luna
Proprietary
License
DeepSeek V4 Pro 0813
Open Weight
GPT-5.6 Luna
Proprietary
Release date
DeepSeek V4 Pro 0813
2026-08-13
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
- GPT-5.6 Luna has the higher public score estimate, 65.6 versus 63.48, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.07788 vs $0.0136. Cache-heavy agent loop: $0.0748 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, DeepSeek V4 Pro 0813 or GPT-5.6 Luna?
GPT-5.6 Luna has the higher public score estimate, 65.6 versus 63.48, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, DeepSeek V4 Pro 0813 or GPT-5.6 Luna?
GPT-5.6 Luna leads the public coding lane, 64.5 to 50.3, with Supported evidence for both models, although the 90% intervals overlap.
Which is better for agentic tasks, DeepSeek V4 Pro 0813 or GPT-5.6 Luna?
The like-for-like agentic tasks row is a practical tie on the public lane, 55 against 55.3, inside the 0.5-point band BenchLM treats as level.
Which costs less, DeepSeek V4 Pro 0813 or GPT-5.6 Luna?
For the stated presets, chat costs $0.0033 on DeepSeek V4 Pro 0813 and $0.0008 on GPT-5.6 Luna; repository review costs $0.07788 and $0.0136; the cache-heavy agent loop costs $0.0748 and $0.02. Costs use the listed standard API rates.
Which has the larger context window, DeepSeek V4 Pro 0813 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 evidence57 rows
Agentic
Terminal-Bench 2.0
DeepSeek V4 Pro 081367.9%SourceGPT-5.6 Luna—Not directly comparable
Terminal-Bench 2.1
DeepSeek V4 Pro 081387.9%SourceGPT-5.6 Luna84.7%SourceDeepSeek V4 Pro 0813 leads this result
BrowseComp
DeepSeek V4 Pro 081383.4%SourceGPT-5.6 Luna83.3%SourceDeepSeek V4 Pro 0813 leads this result
HLE w/ tools
DeepSeek V4 Pro 081360.0%SourceGPT-5.6 Luna—Not directly comparable
MCP Atlas
DeepSeek V4 Pro 081373.6%SourceGPT-5.6 Luna—Not directly comparable
Toolathlon
DeepSeek V4 Pro 081351.8%SourceGPT-5.6 Luna53.4%SourceGPT-5.6 Luna leads this result
CyberGym
DeepSeek V4 Pro 081383.3%SourceGPT-5.6 Luna77.9%SourceDeepSeek V4 Pro 0813 leads this result
Toolathlon-Verified
DeepSeek V4 Pro 081374.1%SourceGPT-5.6 Luna—Not directly comparable
Agents' Last Exam
DeepSeek V4 Pro 081325.7%SourceGPT-5.6 Luna—Not directly comparable
AutomationBench
DeepSeek V4 Pro 081331.8%SourceGPT-5.6 Luna—Not directly comparable
Terminal-Bench 2.1 (Vals)
DeepSeek V4 Pro 081354.7%SourceGPT-5.6 Luna79.0%SourceGPT-5.6 Luna leads this result
Terminal-Bench 3.0
DeepSeek V4 Pro 0813—GPT-5.6 Luna14.3%SourceNot directly comparable
OSWorld 2.0
DeepSeek V4 Pro 0813—GPT-5.6 Luna45.6%SourceNot directly comparable
ExploitGym
DeepSeek V4 Pro 0813—GPT-5.6 Luna12.4%SourceNot directly comparable
ApprenticeBench
DeepSeek V4 Pro 0813—GPT-5.6 Luna7%SourceNot directly comparable
Coding
LiveCodeBench Pass@1-COT
DeepSeek V4 Pro 081393.5%SourceGPT-5.6 Luna—Not directly comparable
Codeforces
DeepSeek V4 Pro 08133206.0SourceGPT-5.6 Luna—Not directly comparable
SWE-bench Verified
DeepSeek V4 Pro 081380.6%SourceGPT-5.6 Luna—Not directly comparable
SWE-bench Pro
DeepSeek V4 Pro 081355.4%SourceGPT-5.6 Luna62.7%SourceGPT-5.6 Luna leads this result
SWE Multilingual
DeepSeek V4 Pro 081376.2%SourceGPT-5.6 Luna—Not directly comparable
Terminal-Bench 2.0
DeepSeek V4 Pro 081367.9%SourceGPT-5.6 Luna—Not directly comparable
Vibe Code Bench
DeepSeek V4 Pro 081349.93%SourceGPT-5.6 Luna—Not directly comparable
Terminal-Bench 2.1
DeepSeek V4 Pro 081387.9%SourceGPT-5.6 Luna84.7%SourceDeepSeek V4 Pro 0813 leads this result
NL2Repo
DeepSeek V4 Pro 081361.5%SourceGPT-5.6 Luna—Not directly comparable
DeepSWE
DeepSeek V4 Pro 081362.7%SourceGPT-5.6 Luna67.2%SourceGPT-5.6 Luna leads this result
DSBench-FullStack
DeepSeek V4 Pro 081371.1%SourceGPT-5.6 Luna—Not directly comparable
DSBench-Hard
DeepSeek V4 Pro 081367.2%SourceGPT-5.6 Luna—Not directly comparable
OpenHarmony Bench
DeepSeek V4 Pro 081359.0%SourceGPT-5.6 Luna—Not directly comparable
LiveCodeBench (Vals)
DeepSeek V4 Pro 081387.5%SourceGPT-5.6 Luna—Not directly comparable
SWE-bench (Vals)
DeepSeek V4 Pro 081396.4%SourceGPT-5.6 Luna93.0%SourceDeepSeek V4 Pro 0813 leads this result
FrontierCode 1.1 Extended
DeepSeek V4 Pro 0813—GPT-5.6 Luna55.1%SourceNot directly comparable
cursorBench32
DeepSeek V4 Pro 0813—GPT-5.6 Luna61.1%SourceNot directly comparable
VulcanBench v3
DeepSeek V4 Pro 0813—GPT-5.6 Luna85.5%SourceNot directly comparable
Reasoning
MRCR 1M
DeepSeek V4 Pro 081383.5%SourceGPT-5.6 Luna—Not directly comparable
CorpusQA 1M
DeepSeek V4 Pro 081362.0%SourceGPT-5.6 Luna—Not directly comparable
ARC-AGI-1
DeepSeek V4 Pro 081390.00%SourceGPT-5.6 Luna—Not directly comparable
ARC-AGI-2
DeepSeek V4 Pro 081361.3%SourceGPT-5.6 Luna59.5%SourceDeepSeek V4 Pro 0813 leads this result
ARC-AGI-3
DeepSeek V4 Pro 0813—GPT-5.6 Luna0.2%SourceNot directly comparable
Multimodal
Knowledge
MMLU-Pro
DeepSeek V4 Pro 081387.5%SourceGPT-5.6 Luna—Not directly comparable
SimpleQA
DeepSeek V4 Pro 081357.9%SourceGPT-5.6 Luna—Not directly comparable
Chinese-SimpleQA
DeepSeek V4 Pro 081384.4%SourceGPT-5.6 Luna—Not directly comparable
GPQA
DeepSeek V4 Pro 081390.1%SourceGPT-5.6 Luna92.3%SourceGPT-5.6 Luna leads this result
GPQA-D
DeepSeek V4 Pro 081390.1%SourceGPT-5.6 Luna92.3%SourceGPT-5.6 Luna leads this result
HLE
DeepSeek V4 Pro 081342.7%SourceGPT-5.6 Luna—Not directly comparable
GPQA Diamond (Vals)
DeepSeek V4 Pro 081392.4%SourceGPT-5.6 Luna91.7%SourceDeepSeek V4 Pro 0813 leads this result
MMLU-Pro (Vals)
DeepSeek V4 Pro 081387.0%SourceGPT-5.6 Luna86.0%SourceDeepSeek V4 Pro 0813 leads this result
HealthBench Professional
DeepSeek V4 Pro 0813—GPT-5.6 Luna55.7%SourceNot directly comparable
HealthBench Hard
DeepSeek V4 Pro 0813—GPT-5.6 Luna32.0%SourceNot directly comparable
Math
HMMT Feb 2026
DeepSeek V4 Pro 081395.2%SourceGPT-5.6 Luna—Not directly comparable
IMOAnswerBench
DeepSeek V4 Pro 081389.8%SourceGPT-5.6 Luna—Not directly comparable
Apex
DeepSeek V4 Pro 081338.3%SourceGPT-5.6 Luna—Not directly comparable
Apex Shortlist
DeepSeek V4 Pro 081390.2%SourceGPT-5.6 Luna—Not directly comparable
FrontierMath (legacy)
DeepSeek V4 Pro 0813—GPT-5.6 Luna78.6%SourceNot directly comparable
FrontierMath v2 (Tiers 1-3)
DeepSeek V4 Pro 0813—GPT-5.6 Luna78.600%SourceNot directly comparable
FrontierMath v2 (Tier 4)
DeepSeek V4 Pro 0813—GPT-5.6 Luna58.500%SourceNot directly comparable
57 public results · 14 shared
Watch DeepSeek V4 Pro 0813 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 27, 2026