DeepSeek V4.1 Flash vs GLM-5.3 vs GPT-5.6 Luna
Too close to call on our weighted score (DeepSeek V4.1 Flash 76, GPT-5.6 Luna 76, GLM-5.3 67). The right pick depends on what you value most.Too close to call on our weighted score (GLM-5.3 81, DeepSeek V4.1 Flash 81, GPT-5.6 Luna 81). The right pick depends on what you value most.
DeepSeek
DeepSeek V4.1 Flash
76/100- ECI157.0
- Price$0.222 / $1.04
- Context1M
Z.ai (Zhipu)
GLM-5.3
67/100- ECI157.1
- Price$1.40 / $4.40
- Context1M
OpenAI
GPT-5.6 Luna
76/100- ECI156.7
- Price$0.20 / $1.20
- Context1.05M
DeepSeek
DeepSeek V4.1 Flash
81/100- ECI157.0
- Price$0.222 / $1.04
- Context1M
Z.ai (Zhipu)
GLM-5.3
81/100- ECI157.1
- Price$1.40 / $4.40
- Context1M
OpenAI
GPT-5.6 Luna
81/100- ECI156.7
- Price$0.20 / $1.20
- Context1.05M
Too close to call
It is close. Our weighted score puts them within a point (DeepSeek V4.1 Flash 76/100, GPT-5.6 Luna 76/100, GLM-5.3 67/100), so choose by what matters most for your work: GLM-5.3 for raw capability, DeepSeek V4.1 Flash on price and GPT-5.6 Luna for long inputs. The score weighs capability 59%, price 29%, context window 12%.It is close. Our weighted score puts them within a point (GLM-5.3 81/100, DeepSeek V4.1 Flash 81/100, GPT-5.6 Luna 81/100), so choose by what matters most for your work: GLM-5.3 for raw capability, DeepSeek V4.1 Flash on price and GPT-5.6 Luna for long inputs. The score weighs capability 83%, context window 17%.
- CapabilityGLM-5.3Capabilities Index (ECI): GLM-5.3 157.1 · DeepSeek V4.1 Flash 157.0 · GPT-5.6 Luna 156.7
- Lowest priceDeepSeek V4.1 FlashDeepSeek V4.1 Flash $0.426 · GPT-5.6 Luna $0.45 · GLM-5.3 $2.15 per 1M tokens (3:1 blend)
- Longest contextGPT-5.6 LunaGPT-5.6 Luna 1,050,000 · DeepSeek V4.1 Flash 1,000,000 · GLM-5.3 1,000,000 tokens
- Widest inputsGPT-5.6 LunaDeepSeek V4.1 Flash: Text, Images · GLM-5.3: Text · GPT-5.6 Luna: Text, Images, PDFs
- Self-hostingDeepSeek V4.1 Flash and GLM-5.3Publishes downloadable weights (MIT)
| Measure | Weight | DeepSeek V4.1 Flash | GLM-5.3 | GPT-5.6 Luna |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 59% | 84 | 85 | 84 |
| Price | 29% | 65 | 33 | 63 |
| Context window | 12% | 64 | 64 | 65 |
| Overall | 100% | 76/100 | 67/100 | 76/100 |
Left out because at least one model lacks the data: inputs & features. The remaining weights were rescaled.
| Measure | Weight | DeepSeek V4.1 Flash | GLM-5.3 | GPT-5.6 Luna |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 83% | 84 | 85 | 84 |
| Context window | 17% | 64 | 64 | 65 |
| Overall | 100% | 81/100 | 81/100 | 81/100 |
Left out because at least one model lacks the data: inputs & features. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Unavailable specifications are marked “Not published”. Publisher results include their evaluation settings and source.
| Specification | |||
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 157.0 | 157.1 (best) | 156.7 |
| ECI rank | #35 of 159 | #34 of 159 (best) | #37 of 159 |
| ECI confidence interval | 155.1–159.2 | 155.0–159.4 | 154.9–158.8 |
| ECI evaluation date | Sep 9, 2026 | Aug 14, 2026 | Jul 9, 2026 |
| Documented capabilities | DeepSeek V4.1 Flash model for reasoning and agentic coding | Flagship GLM model for long-horizon coding, agents, and complex project delivery | Cost-efficient GPT-5.6 model for fast, high-volume workloads |
| Model parameters | Not published | Not published | Not published |
| Output modalities | Text | Text | Text |
| GPQA DiamondGraduate-level science questions | 89.8%max | 90.9%max | 91.6%max (best) |
| FrontierMath Tiers 1–3Research-level mathematics | 67.4%max | 68.8%max | 82.1%max (best) |
| OTIS Mock AIME 2024–2025Competition mathematics | 98.3%max (best) | 91.1%max | 98.3%max (best) |
| SimpleQA VerifiedShort factual questions | 42.9%max (best) | 41.0%max | 41.0%max |
| Publisher-reported benchmarks | |||
| GPQA Diamondreasoning_effort=100 | 90.9pass@1Reported result | Not published | Not published |
| MathArena Apexreasoning_effort=100 | 65.6pass@1Reported result | Not published | Not published |
| Humanity's Last Exam · full setreasoning_effort=100 | 36.8pass@1Reported result | Not published | Not published |
| Humanity's Last Exam · text-only subsetreasoning_effort=100 | 39.1pass@1Reported result | Not published | Not published |
| Codeforcesreasoning_effort=100 | 3471ratingReported result | Not published | Not published |
| Terminal-Bench 2.1reasoning_effort=100 | 90.6pass@1Reported result | Not published | Not published |
| Terminal-Bench 3.0reasoning_effort=100 | 30pass@1Reported result | Not published | Not published |
| Terminal-Bench 4.0reasoning_effort=100 | 31.2pass@1Reported result | Not published | Not published |
| DeepSWE 1.1reasoning_effort=100 | 74.2resolvedReported result | Not published | Not published |
| ProgramBenchreasoning_effort=100 | 20.3almost@1Reported result | Not published | Not published |
| NL2Reporeasoning_effort=100 | 64scoreReported result | Not published | Not published |
| CyberGymreasoning_effort=100 | 88.1pass@1Reported result | Not published | Not published |
| ExploitGymreasoning_effort=100 | 15.3pass@1Reported result | Not published | Not published |
| AutomationBenchreasoning_effort=100 | 54.8pass@1Reported result | Not published | Not published |
| Agents' Last Examreasoning_effort=100 | 31.8pass@1Reported result | Not published | Not published |
| SEC-Bench Proreasoning_effort=100 | 62.8pass@1Reported result | Not published | Not published |
| Humanity's Last Examreasoning_effort=100; with tools | 63.9pass@1Reported result | Not published | Not published |
| Chartographyreasoning_effort=100; with tools | 78.9pass@1Reported result | Not published | Not published |
| BabyVisionreasoning_effort=100; with tools | 89.6pass@1Reported result | Not published | Not published |
| ZeroBench · mainreasoning_effort=100; with tools | 49pass@5Reported result | Not published | Not published |
| Terminal-Bench 2.1max effort | Not published | 88.2pass@1Reported result | Not published |
| Terminal-Bench 3.0max effort | Not published | 28.3avg@3Reported result | Not published |
| DeepSWE 1.1max effort; 6h timeout; 400K context | Not published | 66.9resolvedReported result | Not published |
| NL2Repomax effort | Not published | 58scoreReported result | Not published |
| ProgramBenchmax effort | Not published | 19almost solvedReported result | Not published |
| FrontierSWEmax effort | Not published | 78.1dominance scoreReported resultAug 14, 2026 | Not published |
| SWE-Marathon 1.1max effort; modified anti-cheat checks | Not published | 42.5scoreReported result | Not published |
| PostTrainBenchmax effort; 3 runs; modified external-API checks | Not published | 39.8weighted averageReported result | Not published |
| CyberGymmax effort | Not published | 84.5pass@1Reported result | Not published |
| ExploitGym · 869 tasksmax effort; 2h rescaled timeout | Not published | 105tasks solvedReported result | Not published |
| ExploitGym · 869 tasksmax effort; 6h rescaled timeout | Not published | 130tasks solvedReported result | Not published |
| ExploitBenchmax effort; union over 3 revisions | Not published | 54.4average coverageReported result | Not published |
| Toolathlon-Verifiedmax effort | Not published | 73pass@1Reported result | Not published |
| AutomationBench 1.0.6max effort | Not published | 48.2pass@1Reported result | Not published |
| Agents' Last Exammax effort; CLI | Not published | 28.5scoreReported result | Not published |
| Humanity's Last Exammax effort; with tools | Not published | 62.5accuracyReported result | Not published |
| GDPval-AA 2max effort | Not published | 1769EloReported result | Not published |
| SWE-Bench ProPublisher-reported result | Not published | Not published | 62.7resolve rateReported resultJul 9, 2026 |
| Terminal-Bench 2.1Publisher-reported result | Not published | Not published | 84.7success rateReported resultJul 9, 2026 |
| DeepSWE 1.1Publisher-reported result | Not published | Not published | 67.2resolve rateReported resultJul 9, 2026 |
| GPQA DiamondPublisher-reported result | Not published | Not published | 92.3accuracyReported resultJul 9, 2026 |
| FrontierMath v2 · Tier 1-3Publisher-reported result | Not published | Not published | 78.6accuracyReported resultJul 9, 2026 |
| BrowseCompPublisher-reported result | Not published | Not published | 83.3accuracyReported resultJul 9, 2026 |
| OSWorld 2.0Publisher-reported result | Not published | Not published | 45.6success rateReported resultJul 9, 2026 |
| MMMU Prono tools | Not published | Not published | 78.4accuracyReported resultJul 9, 2026 |
| Agents' Last ExamPublisher-reported result | Not published | Not published | 50.3Reported resultJul 9, 2026 |
| ToolathlonPublisher-reported result | Not published | Not published | 53.4success rateReported resultJul 9, 2026 |
| Artificial Analysis Intelligence Index 4.1max | Not published | Not published | 51.2index scoreReported resultJul 9, 2026 |
| Artificial Analysis Coding Agent Index 1.1max | Not published | Not published | 74.6index scoreReported resultJul 9, 2026 |
| Price per million tokens | |||
| Input | $0.222 | $1.40 | $0.20 (best) |
| Output | $1.04 (best) | $4.40 | $1.20 |
| Cached input | Not published | $0.26 | $0.02 (best) |
| Blended (3:1) | $0.426 (best) | $2.15 | $0.45 |
| Long-context rate | Same rate | Same rate | Over 272K: $0.40 / $1.80 |
| Price source | Median of 82 providers | Official Z.AI API | Official OpenAI API |
| Limits | |||
| Context window | 1,000,000 tokens | 1,000,000 tokens | 1,050,000 tokens (best) |
| Max input | Not published | Not published | 922,000 tokens |
| Max output | 384,000 tokens (best) | 131,072 tokens | 128,000 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | No | Yes |
| PDFs | No | No | Yes |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yes | Yeslow · high · max | Yeslow · medium · high · xhigh · max |
| Tool calling | Yes | Yes | Yes |
| Structured output | Not published | Yes | Yes |
| File attachments | Yes | No | Yes |
| Temperature control | Yes | Yes | Yes |
| Reasoning control | always | effort | effort |
| Availability | |||
| Weights | OpenMIT | Open | Proprietary |
| API model ID | Not published | glm-5.3 | gpt-5.6-luna |
| API providers | 86 | 123 (best) | 57 |
| Released | Sep 10, 2026 | Aug 14, 2026 | Jul 9, 2026 |
| Knowledge cutoff | May 2025 | Not published | Feb 16, 2026 |
| Publisher documentation | Model weights | Publisher documentation · Model weights | Publisher documentation |
| Source metadata updated | Sep 10, 2026 | Aug 14, 2026 | Jul 9, 2026 |
| Data sources | models.dev · Epoch AI | models.dev · Epoch AI | models.dev · Epoch AI |
| Catalog checked | Nov 22, 2026 | Nov 22, 2026 | Nov 22, 2026 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
DeepSeek V4.1 Flash$4.29
GLM-5.3$22.80
GPT-5.6 Luna$4.40
Which should you choose?
Which is better: DeepSeek V4.1 Flash, GLM-5.3 or GPT-5.6 Luna?
It is close. Our weighted score puts them within a point (DeepSeek V4.1 Flash 76/100, GPT-5.6 Luna 76/100, GLM-5.3 67/100), so choose by what matters most for your work: GLM-5.3 for raw capability, DeepSeek V4.1 Flash on price and GPT-5.6 Luna for long inputs. The score weighs capability 59%, price 29%, context window 12%.It is close. Our weighted score puts them within a point (GLM-5.3 81/100, DeepSeek V4.1 Flash 81/100, GPT-5.6 Luna 81/100), so choose by what matters most for your work: GLM-5.3 for raw capability, DeepSeek V4.1 Flash on price and GPT-5.6 Luna for long inputs. The score weighs capability 83%, context window 17%.
Which is cheaper, DeepSeek V4.1 Flash, GLM-5.3 or GPT-5.6 Luna?
DeepSeek V4.1 Flash is cheaper at $0.222 input / $1.04 output per million tokens (median across 82 API providers). GPT-5.6 Luna costs $0.20 input / $1.20 output per million tokens (official OpenAI API price); GLM-5.3 costs $1.40 input / $4.40 output per million tokens (official Z.AI API price). At a typical mix of three input tokens to one output token, that is $0.426 per million tokens for DeepSeek V4.1 Flash versus $0.45 for GPT-5.6 Luna (1.1× as much) and $2.15 for GLM-5.3 (5× as much).
Which scores higher on benchmarks?
GLM-5.3 scores higher on the Capabilities Index (ECI): GLM-5.3 157.1 (#34 of 159), DeepSeek V4.1 Flash 157.0 (#35 of 159) and GPT-5.6 Luna 156.7 (#37 of 159). The confidence ranges of the top two overlap (155.0–159.4 vs 155.1–159.2), so treat the gap as small. On individual benchmarks: GPQA Diamond — GPT-5.6 Luna 91.6%, GLM-5.3 90.9%, DeepSeek V4.1 Flash 89.8%; FrontierMath Tiers 1–3 — GPT-5.6 Luna 82.1%, GLM-5.3 68.8%, DeepSeek V4.1 Flash 67.4%; OTIS Mock AIME 2024–2025 — DeepSeek V4.1 Flash 98.3%, GPT-5.6 Luna 98.3%, GLM-5.3 91.1%; SimpleQA Verified — DeepSeek V4.1 Flash 42.9%, GLM-5.3 41.0%, GPT-5.6 Luna 41.0%.
Which is better for coding?
There are no published SWE-bench Verified results for DeepSeek V4.1 Flash, GLM-5.3 and GPT-5.6 Luna yet, so there is no like-for-like coding score. On overall capability, GLM-5.3 leads, which tends to carry over to coding, but test on your own codebase. All three support tool calling for agent workflows.
Which has the bigger context window?
GPT-5.6 Luna has the largest context window at 1,050,000 tokens, against 1,000,000 for DeepSeek V4.1 Flash and 1,000,000 for GLM-5.3. Maximum output per response: DeepSeek V4.1 Flash up to 384,000, GLM-5.3 up to 131,072, GPT-5.6 Luna up to 128,000 tokens.
Which can read images, PDFs, audio or video?
DeepSeek V4.1 Flash accepts text and images; GLM-5.3 accepts text; GPT-5.6 Luna accepts text, images and PDFs. GPT-5.6 Luna handles the widest range of inputs.
Are any of these open source?
DeepSeek V4.1 Flash and GLM-5.3 publishes its weights (MIT) and can be self-hosted; GPT-5.6 Luna is proprietary.
Which is newer?
DeepSeek V4.1 Flash is the newest, released Sep 10, 2026. GLM-5.3 came out Aug 14, 2026; GPT-5.6 Luna came out Jul 9, 2026. Knowledge cutoff: DeepSeek V4.1 Flash May 2025, GPT-5.6 Luna Feb 16, 2026.
How do you decide the winner?
This rating uses capability (59%, independent benchmark results); price (29%, blended price per million tokens (3 input : 1 output), log scale); context window (12%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Nov 22, 2026.This rating uses capability (83%, independent benchmark results); context window (17%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Nov 22, 2026.