Coding
Like-for-like- GPT-5.6 Luna
- 62.7
- Hy4 preview
- 65.7
- Weighted basis
- 1 vs 1 rows
- Reading
- Hy4 preview leads
GPT-5.6 Luna vs Hy4 preview
Updated August 28, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
Hy4 preview has the higher public score estimate, 79.16 versus 67.35, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
5 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.
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.
Coding work
Code generation, repair, and software-engineering tasks
Hy4 preview
Hy4 preview leads on the same 1 weighted benchmark row.
Confidence: limited
Long 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
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
No shared weighted benchmark basis supports a winner.
Confidence: limited
Chat turn cost
1K fresh input + 500 output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Repository review cost
50K fresh input + 3K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
19 GPT-5.6 Luna only
5 Shared
23 Hy4 preview only
- Shared results
- 5
- GPT-5.6 Luna only
- 19
- Hy4 preview only
- 23
- Like-for-like categories
- 1 / 8
1 category uses different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Category results, on a stated basis
Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.
Knowledge
Directional only- GPT-5.6 Luna
- 92.3
- Hy4 preview
- 60.4
- Weighted basis
- 1 vs 2 rows
- Reading
- Directional only
Agentic
Not comparable- GPT-5.6 Luna
- 84.1
- Hy4 preview
- Not measured
- Weighted basis
- 2 vs 0 rows
- Reading
- Not comparable
Reasoning
Not comparable- GPT-5.6 Luna
- 59.5
- Hy4 preview
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
Math
Not comparable- GPT-5.6 Luna
- 73.6
- Hy4 preview
- Not measured
- Weighted basis
- 2 vs 0 rows
- Reading
- Not comparable
Multilingual
Not comparable- GPT-5.6 Luna
- Not measured
- Hy4 preview
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Multimodal
Not comparable- GPT-5.6 Luna
- 78.4
- Hy4 preview
- 66.2
- Weighted basis
- 1 vs 1 rows
- Reading
- Not comparable
Instruction following
Not comparable- GPT-5.6 Luna
- Not measured
- Hy4 preview
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
| Category | GPT-5.6 Luna | Hy4 preview | Weighted basis | Reading |
|---|---|---|---|---|
| Coding | 62.7 | 65.7 | Like-for-like1 vs 1 rows | Hy4 preview leads |
| Knowledge | 92.3 | 60.4 | Directional only1 vs 2 rows | Directional only |
| Agentic | 84.1 | Not measured | Not comparable2 vs 0 rows | Not comparable |
| Reasoning | 59.5 | Not measured | Not comparable1 vs 0 rows | Not comparable |
| Math | 73.6 | Not measured | Not comparable2 vs 0 rows | Not comparable |
| Multilingual | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Multimodal | 78.4 | 66.2 | Not comparable1 vs 1 rows | Not comparable |
| Instruction following | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
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 Pro
Coding
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
- GPT-5.6 Luna
- $0.004
- Fits in one request
- Hy4 preview
- Self-hosted; infrastructure cost varies
- Fits in one request
Hy4 preview has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- GPT-5.6 Luna
- $0.068
- Fits in one request
- Hy4 preview
- Self-hosted; infrastructure cost varies
- Fits in one request
Hy4 preview has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- GPT-5.6 Luna
- $0.1
- Fits in one request
- Hy4 preview
- Self-hosted; infrastructure cost varies
- Fits in one request
- Cached-input rate unavailable
Hy4 preview has no comparable published API token rate.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationGPT-5.6 LunaHy4 preview
Context window
Maximum documented context; output-token limits may be lower.
GPT-5.6 Luna
1.05M
OpenAI model catalogHy4 preview
API model ID
GPT-5.6 Luna
gpt-5.6-luna
OpenAI model catalogHy4 preview
tencent/Hy4-preview
Tencent Hy4 preview model cardCached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
GPT-5.6 Luna
$0.1 per 1M cached input tokens
OpenAI API pricingHy4 preview
No comparable hosted API rate
Tencent Hy4 preview model cardDocumented inputs
GPT-5.6 Luna
text, image
OpenAI model catalogHy4 preview
Not sourced
Documented outputs
GPT-5.6 Luna
Hy4 preview
Not sourced
Provider availability
GPT-5.6 Luna
Generally Available · OpenAI Responses API
OpenAI model catalogHy4 preview
Not sourced
Reasoning profile
GPT-5.6 Luna
Reasoning
Hy4 preview
Reasoning
Weight access
GPT-5.6 Luna
Proprietary
Hy4 preview
Open Weight
License
GPT-5.6 Luna
Proprietary
Hy4 preview
Open Weight
Release date
GPT-5.6 Luna
2026-07-09
Hy4 preview
2026-08-28
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
- Hy4 preview has the higher public score estimate, 79.16 versus 67.35, but the 90% score intervals overlap.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- 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.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence47 rows
Agentic
Terminal-Bench 3.0
GPT-5.6 Luna14.3%SourceHy4 preview—Not directly comparable
Terminal-Bench 2.0
GPT-5.6 Luna84.7%SourceHy4 preview—Not directly comparable
BrowseComp
GPT-5.6 Luna83.3%SourceHy4 preview—Not directly comparable
OSWorld 2.0
GPT-5.6 Luna45.6%SourceHy4 preview—Not directly comparable
CyberGym
GPT-5.6 Luna77.9%SourceHy4 preview78.4%SourceHy4 preview leads this result
ExploitGym
GPT-5.6 Luna12.4%SourceHy4 preview—Not directly comparable
Toolathlon
GPT-5.6 Luna53.4%SourceHy4 preview—Not directly comparable
Terminal-Bench 2.1
GPT-5.6 Luna—Hy4 preview85.4%SourceNot directly comparable
WideResearch
GPT-5.6 Luna—Hy4 preview83.9%SourceNot directly comparable
DRACO
GPT-5.6 Luna—Hy4 preview77.2%SourceNot directly comparable
MCP Atlas
GPT-5.6 Luna—Hy4 preview83.7%SourceNot directly comparable
Toolathlon-Verified
GPT-5.6 Luna—Hy4 preview74.1%SourceNot directly comparable
APEX-Agents
GPT-5.6 Luna—Hy4 preview37.1%SourceNot directly comparable
skillsBench
GPT-5.6 Luna—Hy4 preview62.9%SourceNot directly comparable
JobBench
GPT-5.6 Luna—Hy4 preview61.7%SourceNot directly comparable
Agents' Last Exam
GPT-5.6 Luna—Hy4 preview22.8%SourceNot directly comparable
AutomationBench
GPT-5.6 Luna—Hy4 preview32.1%SourceNot directly comparable
BankerToolBench
GPT-5.6 Luna—Hy4 preview78.6%SourceNot directly comparable
HLE w/ tools
GPT-5.6 Luna—Hy4 preview55.4%SourceNot directly comparable
Coding
SWE-bench Pro
GPT-5.6 Luna62.7%SourceHy4 preview65.7%SourceHy4 preview leads this result
Terminal-Bench 2.0
GPT-5.6 Luna84.7%SourceHy4 preview—Not directly comparable
deepSwe
GPT-5.6 Luna67.2%SourceHy4 preview64.3%SourceGPT-5.6 Luna leads this result
FrontierCode 1.1 Extended
GPT-5.6 Luna55.1%SourceHy4 preview—Not directly comparable
cursorBench32
GPT-5.6 Luna61.1%SourceHy4 preview—Not directly comparable
VulcanBench v3
GPT-5.6 Luna85.5%SourceHy4 preview—Not directly comparable
Terminal-Bench 2.1
GPT-5.6 Luna—Hy4 preview85.4%SourceNot directly comparable
SWE Multilingual
GPT-5.6 Luna—Hy4 preview82.9%SourceNot directly comparable
NL2Repo
GPT-5.6 Luna—Hy4 preview58.9%SourceNot directly comparable
ProgramBench
GPT-5.6 Luna—Hy4 preview17.5%SourceNot directly comparable
PostTrain Bench
GPT-5.6 Luna—Hy4 preview35.6%SourceNot directly comparable
sweMarathon
GPT-5.6 Luna—Hy4 preview31.9%SourceNot directly comparable
Reasoning
Knowledge
GPQA
GPT-5.6 Luna92.3%SourceHy4 preview92.3%SourceTie
GPQA-D
GPT-5.6 Luna92.3%SourceHy4 preview92.3%SourceTie
HealthBench Professional
GPT-5.6 Luna55.7%SourceHy4 preview—Not directly comparable
HealthBench Hard
GPT-5.6 Luna32.0%SourceHy4 preview—Not directly comparable
HLE
GPT-5.6 Luna—Hy4 preview55.4%SourceNot directly comparable
HLE w/o tools
GPT-5.6 Luna—Hy4 preview43.4%SourceNot directly comparable
Math
FrontierMath (legacy)
GPT-5.6 Luna78.6%SourceHy4 preview—Not directly comparable
FrontierMath v2 (Tiers 1-3)
GPT-5.6 Luna78.600%SourceHy4 preview—Not directly comparable
FrontierMath v2 (Tier 4)
GPT-5.6 Luna58.500%SourceHy4 preview—Not directly comparable
Apex
GPT-5.6 Luna—Hy4 preview74.2%SourceNot directly comparable
Multimodal
Frequently asked questions
Which is better, GPT-5.6 Luna or Hy4 preview?
Hy4 preview has the higher public score estimate, 79.16 versus 67.35, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, GPT-5.6 Luna or Hy4 preview?
Hy4 preview leads the like-for-like coding comparison across 1 shared weighted benchmark row.
Which is better for agentic tasks, GPT-5.6 Luna or Hy4 preview?
The published evidence does not provide a shared weighted agentic tasks basis for both models, so BenchLM does not name a agentic tasks winner.
Which costs less, GPT-5.6 Luna or Hy4 preview?
Both models do not have comparable published API token rates, so this page does not name a universal price winner.
Which has the larger context window, GPT-5.6 Luna or Hy4 preview?
GPT-5.6 Luna has the larger documented context window: 1.05M, compared with 1M.
Related comparisons
Last updated August 28, 2026
Watch GPT-5.6 Luna vs Hy4 preview
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.