Agentic
Not comparable- GPT-5.6 Luna
- 56.8
- Supported · #32/152
- K-EXAONE 2.0
- Not ranked
- Basis
- BenchAlign lane · 8 vs 2 public rows
- Reading
- Not comparable
GPT-5.6 Luna vs K-EXAONE 2.0
Updated September 10, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
1 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
Share or export
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: documented
Show secondary and unsupported calls
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
K-EXAONE 2.0 is not ranked on the public lane for coding, so no winner is named for coding.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
K-EXAONE 2.0 is not ranked on the public lane for agentic, so no winner is named for agentic.
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.
27 GPT-5.6 Luna only
1 Shared
14 K-EXAONE 2.0 only
- Shared results
- 1
- GPT-5.6 Luna only
- 27
- K-EXAONE 2.0 only
- 14
- Like-for-like categories
- 0 / 8
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign 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.
Coding
Not comparable- GPT-5.6 Luna
- 66.8
- Supported · #9/151
- K-EXAONE 2.0
- Not ranked
- Basis
- BenchAlign lane · 7 vs 3 public rows
- Reading
- Not comparable
Reasoning
Not comparable- GPT-5.6 Luna
- 51.2
- #19/20
- K-EXAONE 2.0
- Not ranked
- Basis
- Provisional lane · 2 vs 0 weighted rows
- Reading
- Not comparable
Knowledge
Not comparable- GPT-5.6 Luna
- 64.1
- Supported · #27/183
- K-EXAONE 2.0
- Not ranked
- Basis
- BenchAlign lane · 6 vs 4 public rows
- Reading
- Not comparable
Math
Not comparable- GPT-5.6 Luna
- 94.4
- Unranked · 3 rankable rows
- K-EXAONE 2.0
- Not ranked
- Basis
- Provisional lane · 2 vs 2 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- GPT-5.6 Luna
- Not ranked
- K-EXAONE 2.0
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- GPT-5.6 Luna
- 67.1
- #21/48
- K-EXAONE 2.0
- Not ranked
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- GPT-5.6 Luna
- Not ranked
- K-EXAONE 2.0
- Not ranked
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
| Category | GPT-5.6 Luna | K-EXAONE 2.0 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 56.8Supported · #32/152 | Not ranked | Not comparableBenchAlign lane · 8 vs 2 public rows | Not comparable |
| Coding | 66.8Supported · #9/151 | Not ranked | Not comparableBenchAlign lane · 7 vs 3 public rows | Not comparable |
| Reasoning | 51.2#19/20 | Not ranked | Not comparableProvisional lane · 2 vs 0 weighted rows | Not comparable |
| Knowledge | 64.1Supported · #27/183 | Not ranked | Not comparableBenchAlign lane · 6 vs 4 public rows | Not comparable |
| Math | 94.4Unranked · 3 rankable rows | Not ranked | Not comparableProvisional lane · 2 vs 2 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multimodal | 67.1#21/48 | Not ranked | Not comparableProvisional lane · 1 vs 0 weighted rows | Not comparable |
| Instruction following | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign) differ from the provisional-lane categories. Unranked scores sit on the lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
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.
A shared-evidence shape is not available.
BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.
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
- K-EXAONE 2.0
- Self-hosted; infrastructure cost varies
- Fits in one request
K-EXAONE 2.0 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
- K-EXAONE 2.0
- Self-hosted; infrastructure cost varies
- Fits in one request
K-EXAONE 2.0 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
- K-EXAONE 2.0
- Self-hosted; infrastructure cost varies
- Fits in one request
- Cached-input rate unavailable
K-EXAONE 2.0 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 LunaK-EXAONE 2.0
Context window
Maximum documented context; output-token limits may be lower.
GPT-5.6 Luna
1.05M
OpenAI model catalogK-EXAONE 2.0
API model ID
GPT-5.6 Luna
gpt-5.6-luna
OpenAI model catalogK-EXAONE 2.0
LGAI-EXAONE/K-EXAONE-2.0-750B-A37B
LG AI Research K-EXAONE 2.0 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 pricingK-EXAONE 2.0
No comparable hosted API rate
LG AI Research K-EXAONE 2.0 model cardDocumented inputs
GPT-5.6 Luna
text, image
OpenAI model catalogK-EXAONE 2.0
Not sourced
Documented outputs
GPT-5.6 Luna
K-EXAONE 2.0
Not sourced
Provider availability
GPT-5.6 Luna
Generally Available · OpenAI Responses API
OpenAI model catalogK-EXAONE 2.0
Not sourced
Reasoning profile
GPT-5.6 Luna
Reasoning
K-EXAONE 2.0
Reasoning
Weight access
GPT-5.6 Luna
Proprietary
K-EXAONE 2.0
Open Weight
License
GPT-5.6 Luna
Proprietary
K-EXAONE 2.0
Open Weight
Release date
GPT-5.6 Luna
2026-07-09
K-EXAONE 2.0
2026-07-31
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
- At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
- 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 evidence42 rows
Agentic
Terminal-Bench 3.0
GPT-5.6 Luna14.3%SourceK-EXAONE 2.0—Not directly comparable
Terminal-Bench 2.0
GPT-5.6 Luna84.7%SourceK-EXAONE 2.0—Not directly comparable
BrowseComp
GPT-5.6 Luna83.3%SourceK-EXAONE 2.0—Not directly comparable
OSWorld 2.0
GPT-5.6 Luna45.6%SourceK-EXAONE 2.0—Not directly comparable
CyberGym
GPT-5.6 Luna77.9%SourceK-EXAONE 2.0—Not directly comparable
ExploitGym
GPT-5.6 Luna12.4%SourceK-EXAONE 2.0—Not directly comparable
Toolathlon
GPT-5.6 Luna53.4%SourceK-EXAONE 2.0—Not directly comparable
Terminal-Bench 2.1 (Vals)
GPT-5.6 Luna79.0%SourceK-EXAONE 2.0—Not directly comparable
Terminal-Bench 2.1
GPT-5.6 Luna—K-EXAONE 2.043.8%SourceNot directly comparable
Claw-Eval
GPT-5.6 Luna—K-EXAONE 2.077.7%SourceNot directly comparable
Coding
SWE-bench Pro
GPT-5.6 Luna62.7%SourceK-EXAONE 2.0—Not directly comparable
Terminal-Bench 2.0
GPT-5.6 Luna84.7%SourceK-EXAONE 2.0—Not directly comparable
DeepSWE
GPT-5.6 Luna67.2%SourceK-EXAONE 2.0—Not directly comparable
FrontierCode 1.1 Extended
GPT-5.6 Luna55.1%SourceK-EXAONE 2.0—Not directly comparable
cursorBench32
GPT-5.6 Luna61.1%SourceK-EXAONE 2.0—Not directly comparable
VulcanBench v3
GPT-5.6 Luna85.5%SourceK-EXAONE 2.0—Not directly comparable
SWE-bench (Vals)
GPT-5.6 Luna93.0%SourceK-EXAONE 2.0—Not directly comparable
SciCode
GPT-5.6 Luna—K-EXAONE 2.037.4%SourceNot directly comparable
SWE-bench Verified
GPT-5.6 Luna—K-EXAONE 2.068.2%SourceNot directly comparable
Terminal-Bench 2.1
GPT-5.6 Luna—K-EXAONE 2.043.8%SourceNot directly comparable
Reasoning
Knowledge
GPQA
GPT-5.6 Luna92.3%SourceK-EXAONE 2.0—Not directly comparable
GPQA-D
GPT-5.6 Luna92.3%SourceK-EXAONE 2.082.2%SourceGPT-5.6 Luna leads this result
HealthBench Professional
GPT-5.6 Luna55.7%SourceK-EXAONE 2.0—Not directly comparable
HealthBench Hard
GPT-5.6 Luna32.0%SourceK-EXAONE 2.0—Not directly comparable
GPQA Diamond (Vals)
GPT-5.6 Luna91.7%SourceK-EXAONE 2.0—Not directly comparable
MMLU-Pro (Vals)
GPT-5.6 Luna86.0%SourceK-EXAONE 2.0—Not directly comparable
MMLU-Pro
GPT-5.6 Luna—K-EXAONE 2.083.5%SourceNot directly comparable
HLE
GPT-5.6 Luna—K-EXAONE 2.018.3%SourceNot directly comparable
MMMLU
GPT-5.6 Luna—K-EXAONE 2.086.6%SourceNot directly comparable
Math
FrontierMath (legacy)
GPT-5.6 Luna78.6%SourceK-EXAONE 2.0—Not directly comparable
FrontierMath v2 (Tiers 1-3)
GPT-5.6 Luna78.600%SourceK-EXAONE 2.0—Not directly comparable
FrontierMath v2 (Tier 4)
GPT-5.6 Luna58.500%SourceK-EXAONE 2.0—Not directly comparable
AIME26
GPT-5.6 Luna—K-EXAONE 2.092.3%SourceNot directly comparable
HMMT Feb 2026
GPT-5.6 Luna—K-EXAONE 2.078.4%SourceNot directly comparable
IMOAnswerBench
GPT-5.6 Luna—K-EXAONE 2.078.6%SourceNot directly comparable
Multilingual
PolyMath
GPT-5.6 Luna—K-EXAONE 2.071.3%SourceNot directly comparable
Multimodal
Instruction following
Frequently asked questions
Which is better, GPT-5.6 Luna or K-EXAONE 2.0?
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. The page therefore keeps the decision tied to the specific documented workload.
Which is better for coding, GPT-5.6 Luna or K-EXAONE 2.0?
K-EXAONE 2.0 is not ranked on the public lane for coding, so no winner is named for coding.
Which is better for agentic tasks, GPT-5.6 Luna or K-EXAONE 2.0?
K-EXAONE 2.0 is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.
Which costs less, GPT-5.6 Luna or K-EXAONE 2.0?
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 K-EXAONE 2.0?
GPT-5.6 Luna has the larger documented context window: 1.05M, compared with 262K.
Related comparisons
Last updated September 10, 2026
Watch GPT-5.6 Luna vs K-EXAONE 2.0
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.