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
GLM-5 vs GPT-5.6 Luna
Updated November 10, 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.
Share or export
Decision reading
GPT-5.6 Luna has the higher public score estimate, 63.72 versus 54.43, 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.
- API price per 1M tokens
GLM-5: $1 input / $3.20 output
GPT-5.6 Luna: $0.20 input / $1.20 output
- Context window
GLM-5: 200K
GPT-5.6 Luna: 1.05M
- Released
GLM-5: February 12, 2026
GPT-5.6 Luna: July 9, 2026
- Shared results
- 6
- GLM-5 only
- 32
- GPT-5.6 Luna only
- 29
- Like-for-like categories
- 1 / 8
Supported: GLM-5 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.
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-rates
Show secondary and unsupported calls
Repository 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-ratesCoding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
GLM-5 is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.
Confidence: limitedAgentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
GLM-5 is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.
Confidence: limitedCache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. GLM-5 does not fit this workload in one request. GLM-5 has no published cached-input rate, so cached tokens use its listed input rate.
Confidence: rate-fallback
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.
39.7GLM-561.9GPT-5.6 Luna
Directional only · BenchAlign v5.8
GPT-5.6 Luna has the higher coding point estimate. Conditional score ranges do not establish rank confidence.
Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.
Same basis rules as the category table below
Or describe your own job
The LLM Selector reads what you need a model for and returns a shortlist, ranked in the same order the job pages publish.
Describe the task and constraints. This selector recommends models; it cannot carry out the task or access your files and accounts.
We keep what you type here, drafts included, for 180 days, and after that only the job and requirements read from it. Nothing is tied to your IP address or account. Leave out names and other personal details. Privacy
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
32 GLM-5 only
6 Shared
29 GPT-5.6 Luna only
3 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.
BrowseCompAgentic
Normalized gap 21.3GLM-5:62%GPT-5.6 Luna:83.3%SWE-bench ProCoding
Normalized gap 7.6GLM-5:55.1%GPT-5.6 Luna:62.7%GPQAKnowledge
Normalized gap 6.3GLM-5:86%GPT-5.6 Luna:92.3%
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 v5.8 lane for agentic, coding, reasoning, multimodal, 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.
Reasoning
Like-for-like- GLM-5
- 42.4
- Supported · #76/158
- GPT-5.6 Luna
- 58.3
- Supported · #36/158
- Basis
- BenchAlign v5.8 lane · 2 vs 3 public rows
- Reading
- GPT-5.6 Luna leads · intervals overlap
Agentic
Directional only- GLM-5
- 43.1
- Estimated · #67/132
- GPT-5.6 Luna
- 59.1
- Supported · #33/132
- Basis
- BenchAlign v5.8 lane · 13 vs 11 public rows
- Reading
- Directional only
Coding
Directional only- GLM-5
- 39.7
- Estimated · #74/158
- GPT-5.6 Luna
- 61.9
- Supported · #17/158
- Basis
- BenchAlign v5.8 lane · 8 vs 8 public rows
- Reading
- Directional only
Knowledge
Directional only- GLM-5
- 49.5
- Estimated · #82/179
- GPT-5.6 Luna
- 63.1
- Supported · #40/179
- Basis
- BenchAlign v5.8 lane · 6 vs 6 public rows
- Reading
- Directional only
Multimodal
Not comparable- GLM-5
- Not ranked
- GPT-5.6 Luna
- 65.4
- Supported · #23/67
- Basis
- BenchAlign v5.8 lane · 0 vs 4 public rows
- Reading
- Not comparable
Multilingual
Not comparable- GLM-5
- 89.5
- #6/17
- GPT-5.6 Luna
- Not ranked
- Basis
- Provisional lane · 1 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- GLM-5
- 88.6
- #38/157
- GPT-5.6 Luna
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- GLM-5
- 72.5
- Unranked · 6 rankable rows
- GPT-5.6 Luna
- 78.9
- Unranked · 3 rankable rows
- Basis
- Provisional lane · 2 vs 2 weighted rows
- Reading
- Not comparable
| Category | GLM-5 | GPT-5.6 Luna | Basis | Reading |
|---|---|---|---|---|
| Reasoning | 42.4Supported · #76/158 | 58.3Supported · #36/158 | Like-for-likeBenchAlign v5.8 lane · 2 vs 3 public rows | GPT-5.6 Luna leads · intervals overlap |
| Agentic | 43.1Estimated · #67/132 | 59.1Supported · #33/132 | Directional onlyBenchAlign v5.8 lane · 13 vs 11 public rows | Directional only |
| Coding | 39.7Estimated · #74/158 | 61.9Supported · #17/158 | Directional onlyBenchAlign v5.8 lane · 8 vs 8 public rows | Directional only |
| Knowledge | 49.5Estimated · #82/179 | 63.1Supported · #40/179 | Directional onlyBenchAlign v5.8 lane · 6 vs 6 public rows | Directional only |
| Multimodal | Not ranked | 65.4Supported · #23/67 | Not comparableBenchAlign v5.8 lane · 0 vs 4 public rows | Not comparable |
| Multilingual | 89.5#6/17 | Not ranked | Not comparableProvisional lane · 1 vs 0 weighted rows | Not comparable |
| Instruction following | 88.6#38/157 | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | 72.5Unranked · 6 rankable rows | 78.9Unranked · 3 rankable rows | Not comparableProvisional lane · 2 vs 2 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, reasoning, multimodal, and knowledge denominators (BenchAlign v5.8) 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
- GLM-5
- $0.0026
- 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
- GLM-5
- $0.0596
- 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
- GLM-5
- $0.252
- Does not fit in one request
- Cached input priced at the published list-input rate
- GPT-5.6 Luna
- $0.02
- Fits in one request
GLM-5 does not fit this workload in one request. GLM-5 has no published cached-input rate, so cached tokens use its listed input rate.
Your workload
Enter a month of tokens to price it at both models’ listed API rates.
- GLM-5
- $16.40
- per month
- GPT-5.6 Luna
- $4.40
- per month
Standard listed rates, no batch or volume discounts.
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.
SpecificationGLM-5GPT-5.6 Luna
Context window
Maximum documented context; output-token limits may be lower.
GLM-5
200K
GPT-5.6 Luna
1.05M
OpenAI model catalogAPI model ID
GLM-5
Not sourced
GPT-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.
GLM-5
Not published
GPT-5.6 Luna
$0.02 per 1M cached input tokens
OpenAI pricingDocumented inputs
GLM-5
text
Z.AI GLM-5 guideGPT-5.6 Luna
text, image
OpenAI model catalogDocumented outputs
GLM-5
text
Z.AI GLM-5 guideGPT-5.6 Luna
Provider availability
GLM-5
Not sourced
GPT-5.6 Luna
Generally Available · OpenAI Responses API
OpenAI model catalogReasoning profile
GLM-5
Non-Reasoning
GPT-5.6 Luna
Reasoning
Weight access
GLM-5
Open Weight
GPT-5.6 Luna
Proprietary
License
GLM-5
Open Weight
GPT-5.6 Luna
Proprietary
Release date
GLM-5
2026-02-12
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, 63.72 versus 54.43, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.0596 vs $0.0136. Cache-heavy agent loop: $0.252 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, GLM-5 or GPT-5.6 Luna?
GPT-5.6 Luna has the higher public score estimate, 63.72 versus 54.43, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, GLM-5 or GPT-5.6 Luna?
GPT-5.6 Luna scores higher for coding on the public lane, 61.9 to 39.7. GLM-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, GLM-5 or GPT-5.6 Luna?
GPT-5.6 Luna scores higher for agentic tasks on the public lane, 59.1 to 43.1. GLM-5 is scored on Estimated evidence for agentic tasks, so the reading is directional rather than like-for-like. BenchLM does not name a winner for agentic tasks from a directional reading; read the shared benchmark rows directly and test the models on the same task set.
Which costs less, GLM-5 or GPT-5.6 Luna?
For the stated presets, chat costs $0.0026 on GLM-5 and $0.0008 on GPT-5.6 Luna; repository review costs $0.0596 and $0.0136; the cache-heavy agent loop costs $0.252 and $0.02. GLM-5 does not fit this workload in one request. GLM-5 has no published cached-input rate, so cached tokens use its listed input rate.
Which has the larger context window, GLM-5 or GPT-5.6 Luna?
GPT-5.6 Luna has the larger documented context window: 1.05M, compared with 200K.
Self-host vs API cost
Estimates at 50,000 req/day · 1000 tokens/req average.
GLM-5
API / mo$3,150
Self-host / mo$27,962
Break-evenNot reached at 70% load
GPT-5.6 Luna
API / mo$1,050
Self-host / moNot listed
Break-even—
Proprietary model — self-hosting not applicable.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence67 rows
Agentic
Terminal-Bench 2.0
GLM-556.2%SourceGPT-5.6 Luna—Not directly comparable
BrowseComp
GLM-562%SourceGPT-5.6 Luna83.3%SourceGPT-5.6 Luna leads this result
HLE w/ tools
GLM-550.4%SourceGPT-5.6 Luna—Not directly comparable
Claw-Eval
GLM-557.7%SourceGPT-5.6 Luna—Not directly comparable
QwenClawBench
GLM-554.1%SourceGPT-5.6 Luna—Not directly comparable
τ³-bench results
GLM-565.6%SourceGPT-5.6 Luna—Not directly comparable
DeepPlanning
GLM-514.6%SourceGPT-5.6 Luna—Not directly comparable
Toolathlon
GLM-538%SourceGPT-5.6 Luna53.4%SourceGPT-5.6 Luna leads this result
MCP Atlas
GLM-531.1%SourceGPT-5.6 Luna—Not directly comparable
MCP-Tasks
GLM-560.8%SourceGPT-5.6 Luna—Not directly comparable
WideResearch
GLM-569.8%SourceGPT-5.6 Luna—Not directly comparable
CyberGym
GLM-543.2%SourceGPT-5.6 Luna77.9%SourceGPT-5.6 Luna leads this result
Gert Labs
GLM-550.99%SourceGPT-5.6 Luna—Not directly comparable
Terminal-Bench 4.1
GLM-5—GPT-5.6 Luna17.60%SourceNot directly comparable
Terminal-Bench 3.0
GLM-5—GPT-5.6 Luna14.3%SourceNot directly comparable
Terminal-Bench 2.1
GLM-5—GPT-5.6 Luna84.7%SourceNot directly comparable
OSWorld 2.0
GLM-5—GPT-5.6 Luna45.6%SourceNot directly comparable
ExploitGym
GLM-5—GPT-5.6 Luna12.4%SourceNot directly comparable
Terminal-Bench 2.1 (Vals)
GLM-5—GPT-5.6 Luna79.0%SourceNot directly comparable
ApprenticeBench
GLM-5—GPT-5.6 Luna7%SourceNot directly comparable
MazeBench
GLM-5—GPT-5.6 Luna0%SourceNot directly comparable
Coding
SWE-bench Verified
GLM-577.8%SourceGPT-5.6 Luna—Not directly comparable
SWE-bench Verified*
GLM-572.8%SourceGPT-5.6 Luna—Not directly comparable
SWE-bench Pro
GLM-555.1%SourceGPT-5.6 Luna62.7%SourceGPT-5.6 Luna leads this result
SWE Multilingual
GLM-573.3%SourceGPT-5.6 Luna—Not directly comparable
NL2Repo
GLM-535.9%SourceGPT-5.6 Luna—Not directly comparable
SWE-Rebench
GLM-562.8%SourceGPT-5.6 Luna—Not directly comparable
React Native Evals
GLM-574.8%SourceGPT-5.6 Luna—Not directly comparable
Terminal-Bench 2.0
GLM-556.2%SourceGPT-5.6 Luna—Not directly comparable
Terminal-Bench 2.1
GLM-5—GPT-5.6 Luna84.7%SourceNot directly comparable
DeepSWE
GLM-5—GPT-5.6 Luna67.2%SourceNot directly comparable
FrontierCode 1.1 Extended
GLM-5—GPT-5.6 Luna55.1%SourceNot directly comparable
CursorBench 3.2
GLM-5—GPT-5.6 Luna61.1%SourceNot directly comparable
VulcanBench v3
GLM-5—GPT-5.6 Luna85.5%SourceNot directly comparable
SWE-bench (Vals)
GLM-5—GPT-5.6 Luna93.0%SourceNot directly comparable
CursorBench 4.0
GLM-5—GPT-5.6 Luna35.9%SourceNot directly comparable
Reasoning
LongBench v2
GLM-560.8%SourceGPT-5.6 Luna—Not directly comparable
AI-Needle
GLM-563.3%SourceGPT-5.6 Luna—Not directly comparable
ARC-AGI-2
GLM-5—GPT-5.6 Luna59.5%SourceNot directly comparable
ARC-AGI-3
GLM-5—GPT-5.6 Luna0.2%SourceNot directly comparable
ARC-AGI-1
GLM-5—GPT-5.6 Luna88.00%SourceNot directly comparable
Multimodal
Knowledge
GPQA
GLM-586%SourceGPT-5.6 Luna92.3%SourceGPT-5.6 Luna leads this result
GPQA-D
GLM-586.0%SourceGPT-5.6 Luna92.3%SourceGPT-5.6 Luna leads this result
SuperGPQA
GLM-566.8%SourceGPT-5.6 Luna—Not directly comparable
MMLU-Pro
GLM-585.7%SourceGPT-5.6 Luna—Not directly comparable
MMLU-Pro (Arcee)
GLM-585.8%SourceGPT-5.6 Luna—Not directly comparable
HLE
GLM-530.5%SourceGPT-5.6 Luna—Not directly comparable
HealthBench Professional
GLM-5—GPT-5.6 Luna55.7%SourceNot directly comparable
HealthBench Hard
GLM-5—GPT-5.6 Luna32.0%SourceNot directly comparable
GPQA Diamond (Vals)
GLM-5—GPT-5.6 Luna91.7%SourceNot directly comparable
MMLU-Pro (Vals)
GLM-5—GPT-5.6 Luna86.0%SourceNot directly comparable
Multilingual
Instruction following
IFEval
GLM-592.6%SourceGPT-5.6 Luna—Not directly comparable
Math
AIME26
GLM-595.8%SourceGPT-5.6 Luna—Not directly comparable
AIME25 (Arcee)
GLM-593.3%SourceGPT-5.6 Luna—Not directly comparable
HMMT Feb 2025
GLM-597.5%SourceGPT-5.6 Luna—Not directly comparable
HMMT Nov 2025
GLM-596.9%SourceGPT-5.6 Luna—Not directly comparable
HMMT Feb 2026
GLM-586.4%SourceGPT-5.6 Luna—Not directly comparable
MMAnswerBench
GLM-582.5%SourceGPT-5.6 Luna—Not directly comparable
FrontierMath (legacy)
GLM-5—GPT-5.6 Luna78.6%SourceNot directly comparable
FrontierMath v2 (Tiers 1-3)
GLM-5—GPT-5.6 Luna78.600%SourceNot directly comparable
FrontierMath v2 (Tier 4)
GLM-5—GPT-5.6 Luna58.500%SourceNot directly comparable
67 public results · 6 shared
Watch GLM-5 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 7,000+ subscribers.
Last updated November 10, 2026