GLM-5.3 vs GPT-6 Luna vs Muse Spark 1.3
GPT-6 Luna comes out ahead, 80 to 73 and 64 on our weighted score, and it is the cheaper option too.Muse Spark 1.3 comes out ahead, 86 to 81 and 74 on our weighted score, though GPT-6 Luna is 10× cheaper per token.
Z.ai (Zhipu)
GLM-5.3
64/100- ECI157.1
- Price$1.40 / $4.40
- Context1M
- Our pick
OpenAI
GPT-6 Luna
80/100- ECI156.5
- Price$0.10 / $0.50
- Context1.05M
Meta
Muse Spark 1.3
73/100- ECI158.2
- Price$1.25 / $4.25
- Context1.05M
Z.ai (Zhipu)
GLM-5.3
74/100- ECI157.1
- Price$1.40 / $4.40
- Context1M
OpenAI
GPT-6 Luna
81/100- ECI156.5
- Price$0.10 / $0.50
- Context1.05M
- Our pick
Meta
Muse Spark 1.3
86/100- ECI158.2
- Price$1.25 / $4.25
- Context1.05M
GPT-6 Luna is our pickMuse Spark 1.3 is our pick
GPT-6 Luna is the better all-round choice, scoring 80/100 against Muse Spark 1.3 (73) and GLM-5.3 (64). It leads on price. Muse Spark 1.3 wins on inputs & features. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.Muse Spark 1.3 is the better all-round choice, scoring 86/100 against GPT-6 Luna (81) and GLM-5.3 (74). It leads on inputs & features. The score weighs capability 67%, inputs & features 20%, context window 13%.
- CapabilityMuse Spark 1.3Capabilities Index (ECI): Muse Spark 1.3 158.2 · GLM-5.3 157.1 · GPT-6 Luna 156.5
- Lowest priceGPT-6 LunaGPT-6 Luna $0.20 · Muse Spark 1.3 $2.00 · GLM-5.3 $2.15 per 1M tokens (3:1 blend)
- Longest contextGPT-6 Luna and Muse Spark 1.3GPT-6 Luna 1,050,000 · Muse Spark 1.3 1,048,576 · GLM-5.3 1,000,000 tokens
- Widest inputsMuse Spark 1.3GLM-5.3: Text · GPT-6 Luna: Text, Images, PDFs · Muse Spark 1.3: Text, Images, PDFs, Audio, Video
- Self-hostingGLM-5.3Publishes downloadable weights
| Measure | Weight | GLM-5.3 | GPT-6 Luna | Muse Spark 1.3 |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 50% | 85 | 84 | 86 |
| Price | 25% | 33 | 79 | 34 |
| Inputs & features | 15% | 45 | 80 | 100 |
| Context window | 10% | 64 | 65 | 65 |
| Overall | 100% | 64/100 | 80/100 | 73/100 |
| Measure | Weight | GLM-5.3 | GPT-6 Luna | Muse Spark 1.3 |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 67% | 85 | 84 | 86 |
| Inputs & features | 20% | 45 | 80 | 100 |
| Context window | 13% | 64 | 65 | 65 |
| Overall | 100% | 74/100 | 81/100 | 86/100 |
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) | 157.1 | 156.5 | 158.2 (best) |
| ECI rank | #34 of 159 | #38 of 159 | #28 of 159 (best) |
| ECI confidence interval | 155.0–159.4 | 154.6–158.5 | 156.2–160.5 |
| ECI evaluation date | Aug 14, 2026 | Sep 22, 2026 | Sep 2, 2026 |
| Documented capabilities | Flagship GLM model for long-horizon coding, agents, and complex project delivery | OpenAI's most efficient model for focused, high-volume tasks | Muse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows. It improves long-horizon agent collaboration, instruction following, and coding efficiency relative to Muse Spark 1.2. |
| Model parameters | Not published | Not published | Not published |
| Output modalities | Text | Text | Text |
| GPQA DiamondGraduate-level science questions | 90.9%max (best) | 90.5%max | Not published |
| FrontierMath Tiers 1–3Research-level mathematics | 68.8%max | 79.0%max (best) | 74.4%xhigh |
| OTIS Mock AIME 2024–2025Competition mathematics | 91.1%max | 98.9%max | 99.2%xhigh (best) |
| SimpleQA VerifiedShort factual questions | 41.0%max | 41.4%max (best) | Not published |
| Publisher-reported benchmarks | |||
| Terminal-Bench 2.1max effort | 88.2pass@1Reported result | Not published | 88.8pass@1Reported resultSep 2, 2026 |
| Terminal-Bench 3.0max effort | 28.3avg@3Reported result | Not published | Not published |
| DeepSWE 1.1max effort; 6h timeout; 400K context | 66.9resolvedReported result | Not published | Not published |
| NL2Repomax effort | 58scoreReported result | Not published | Not published |
| ProgramBenchmax effort | 19almost solvedReported result | Not published | Not published |
| FrontierSWEmax effort | 78.1dominance scoreReported resultAug 14, 2026 | Not published | Not published |
| SWE-Marathon 1.1max effort; modified anti-cheat checks | 42.5scoreReported result | Not published | Not published |
| PostTrainBenchmax effort; 3 runs; modified external-API checks | 39.8weighted averageReported result | Not published | Not published |
| CyberGymmax effort | 84.5pass@1Reported result | Not published | Not published |
| ExploitGym · 869 tasksmax effort; 2h rescaled timeout | 105tasks solvedReported result | Not published | Not published |
| ExploitGym · 869 tasksmax effort; 6h rescaled timeout | 130tasks solvedReported result | Not published | Not published |
| ExploitBenchmax effort; union over 3 revisions | 54.4average coverageReported result | Not published | Not published |
| Toolathlon-Verifiedmax effort | 73pass@1Reported result | Not published | Not published |
| AutomationBench 1.0.6max effort | 48.2pass@1Reported result | Not published | Not published |
| Agents' Last Exammax effort; CLI | 28.5scoreReported result | Not published | Not published |
| Humanity's Last Exammax effort; with tools | 62.5accuracyReported result | Not published | Not published |
| GDPval-AA 2max effort | 1769EloReported result | Not published | 1754EloReported resultSep 2, 2026 |
| DeepSWE 1.1max effort | Not published | 66.6scoreReported resultSep 22, 2026 | Not published |
| JobBenchmax effort | Not published | Not published | 64.9mean rubric scoreReported resultSep 2, 2026 |
| OSWorld 2.0 · 08.08max effort | Not published | Not published | 66.9mean partial scoreReported resultSep 2, 2026 |
| OSWorld 2.0 · 08.08max effort | Not published | Not published | 32binary completion rateReported resultSep 2, 2026 |
| DeepSearchQAmax effort | Not published | Not published | 90.3F1Reported resultSep 2, 2026 |
| Agentic IF Index · Meta internalmax effort | Not published | Not published | 57.8scoreReported resultSep 2, 2026 |
| AutomationBench · public v3 task setmax effort | Not published | Not published | 49.6pass@1Reported resultSep 2, 2026 |
| MRCR 2 · 8-needle, 256K-512Kmax effort | Not published | Not published | 98.5mean sequence-match ratioReported resultSep 2, 2026 |
| MRCR 2 · 8-needle, 512K-1Mmax effort | Not published | Not published | 98.1mean sequence-match ratioReported resultSep 2, 2026 |
| DeepSWE 1.1max effort | Not published | Not published | 75.4pass@1Reported resultSep 2, 2026 |
| SWE-Atlas Codebase QnAmax effort | Not published | Not published | 59.4pass@1Reported resultSep 2, 2026 |
| Price per million tokens | |||
| Input | $1.40 | $0.10 (best) | $1.25 |
| Output | $4.40 | $0.50 (best) | $4.25 |
| Cached input | $0.26 | $0.01 (best) | $0.15 |
| Blended (3:1) | $2.15 | $0.20 (best) | $2.00 |
| Long-context rate | Same rate | Over 272K: $0.20 / $0.75 | Same rate |
| Price source | Official Z.AI API | Official OpenAI API | Official Meta API |
| Limits | |||
| Context window | 1,000,000 tokens | 1,050,000 tokens (best) | 1,048,576 tokens |
| Max input | Not published | 922,000 tokens | Not published |
| Max output | 131,072 tokens (best) | 128,000 tokens | 131,072 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | Yes | Yes |
| PDFs | No | Yes | Yes |
| Audio | No | No | Yes |
| Video | No | No | Yes |
| Reasoning | Yeslow · high · max | Yeslow · medium · high · xhigh · max | Yesminimal · low · medium · high · xhigh · max |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | Yes | Yes |
| File attachments | No | Yes | Yes |
| Temperature control | Yes | Yes | Yes |
| Reasoning control | effort | effort | effort |
| Availability | |||
| Weights | Open | Proprietary | Proprietary |
| API model ID | glm-5.3 | gpt-6-luna | muse-spark-1.3 |
| API providers | 123 (best) | 49 | 18 |
| Released | Aug 14, 2026 | Sep 22, 2026 | Sep 2, 2026 |
| Knowledge cutoff | Not published | May 18, 2026 | Not published |
| Publisher documentation | Publisher documentation · Model weights | Publisher documentation | Publisher documentation |
| Source metadata updated | Aug 14, 2026 | Sep 22, 2026 | Sep 2, 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.
GLM-5.3$22.80
GPT-6 Luna$2.00
Muse Spark 1.3$21.00
Which should you choose?
Which is better: GLM-5.3, GPT-6 Luna or Muse Spark 1.3?
GPT-6 Luna is the better all-round choice, scoring 80/100 against Muse Spark 1.3 (73) and GLM-5.3 (64). It leads on price. Muse Spark 1.3 wins on inputs & features. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.Muse Spark 1.3 is the better all-round choice, scoring 86/100 against GPT-6 Luna (81) and GLM-5.3 (74). It leads on inputs & features. The score weighs capability 67%, inputs & features 20%, context window 13%.
Which is cheaper, GLM-5.3, GPT-6 Luna or Muse Spark 1.3?
GPT-6 Luna is cheaper at $0.10 input / $0.50 output per million tokens (official OpenAI API price). Muse Spark 1.3 costs $1.25 input / $4.25 output per million tokens (official Meta 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 $0.20 per million tokens for GPT-6 Luna versus $2.00 for Muse Spark 1.3 (10× as much) and $2.15 for GLM-5.3 (11× as much).
Which scores higher on benchmarks?
Muse Spark 1.3 scores higher on the Capabilities Index (ECI): Muse Spark 1.3 158.2 (#28 of 159), GLM-5.3 157.1 (#34 of 159) and GPT-6 Luna 156.5 (#38 of 159). The confidence ranges of the top two overlap (156.2–160.5 vs 155.0–159.4), so treat the gap as small. On individual benchmarks: FrontierMath Tiers 1–3 — GPT-6 Luna 79.0%, Muse Spark 1.3 74.4%, GLM-5.3 68.8%; OTIS Mock AIME 2024–2025 — Muse Spark 1.3 99.2%, GPT-6 Luna 98.9%, GLM-5.3 91.1%.
Which is better for coding?
There are no published SWE-bench Verified results for GLM-5.3, GPT-6 Luna and Muse Spark 1.3 yet, so there is no like-for-like coding score. On overall capability, Muse Spark 1.3 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-6 Luna and Muse Spark 1.3 have the largest context windows (1,050,000 and 1,048,576 tokens), against 1,000,000 for GLM-5.3. Maximum output per response: GLM-5.3 up to 131,072, GPT-6 Luna up to 128,000, Muse Spark 1.3 up to 131,072 tokens.
Which can read images, PDFs, audio or video?
GLM-5.3 accepts text; GPT-6 Luna accepts text, images and PDFs; Muse Spark 1.3 accepts text, images, PDFs, audio and video. Muse Spark 1.3 handles the widest range of inputs.
Are any of these open source?
GLM-5.3 publishes its weights and can be self-hosted; GPT-6 Luna and Muse Spark 1.3 is proprietary.
Which is newer?
GPT-6 Luna is the newest, released Sep 22, 2026. Muse Spark 1.3 came out Sep 2, 2026; GLM-5.3 came out Aug 14, 2026. Knowledge cutoff: GPT-6 Luna May 18, 2026.
How do you decide the winner?
This rating uses capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, 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 (67%, independent benchmark results); inputs & features (20%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (13%, 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.