GPT-6 Luna vs Muse Spark 1.3 vs DeepSeek V4 Flash 0731
Too close to call on our weighted score (GPT-6 Luna 80, DeepSeek V4 Flash 0731 80, Muse Spark 1.3 68). The right pick depends on what you value most.Too close to call on our weighted score (Muse Spark 1.3 82, GPT-6 Luna 81, DeepSeek V4 Flash 0731 79). The right pick depends on what you value most.
OpenAI
GPT-6 Luna
80/100- ECI156.5
- Price$0.10 / $0.50
- Context1.05M
Meta
Muse Spark 1.3
68/100- ECI158.2
- Price$1.25 / $4.25
- Context1.05M
DeepSeek
DeepSeek V4 Flash 0731
80/100- ECI154.4
- Price$0.14 / $0.30
- Context1M
OpenAI
GPT-6 Luna
81/100- ECI156.5
- Price$0.10 / $0.50
- Context1.05M
Meta
Muse Spark 1.3
82/100- ECI158.2
- Price$1.25 / $4.25
- Context1.05M
DeepSeek
DeepSeek V4 Flash 0731
79/100- ECI154.4
- Price$0.14 / $0.30
- Context1M
Too close to call
It is close. Our weighted score puts them within a point (GPT-6 Luna 80/100, DeepSeek V4 Flash 0731 80/100, Muse Spark 1.3 68/100), so choose by what matters most for your work: Muse Spark 1.3 for raw capability and DeepSeek V4 Flash 0731 on price. The score weighs capability 59%, price 29%, context window 12%.It is close. Our weighted score puts them within 1 points (Muse Spark 1.3 82/100, GPT-6 Luna 81/100, DeepSeek V4 Flash 0731 79/100), so choose by what matters most for your work: Muse Spark 1.3 for raw capability and DeepSeek V4 Flash 0731 on price. The score weighs capability 83%, context window 17%.
- CapabilityMuse Spark 1.3Capabilities Index (ECI): Muse Spark 1.3 158.2 · GPT-6 Luna 156.5 · DeepSeek V4 Flash 0731 154.4
- Lowest priceDeepSeek V4 Flash 0731DeepSeek V4 Flash 0731 $0.18 · GPT-6 Luna $0.20 · Muse Spark 1.3 $2.00 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 · DeepSeek V4 Flash 0731 1,000,000 tokens
- Widest inputsMuse Spark 1.3GPT-6 Luna: Text, Images, PDFs · Muse Spark 1.3: Text, Images, PDFs, Audio, Video · DeepSeek V4 Flash 0731: Text
- Self-hostingDeepSeek V4 Flash 0731Publishes downloadable weights (MIT)
| Measure | Weight | GPT-6 Luna | Muse Spark 1.3 | DeepSeek V4 Flash 0731 |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 59% | 84 | 86 | 82 |
| Price | 29% | 79 | 34 | 81 |
| Context window | 12% | 65 | 65 | 64 |
| Overall | 100% | 80/100 | 68/100 | 80/100 |
Left out because at least one model lacks the data: inputs & features. The remaining weights were rescaled.
| Measure | Weight | GPT-6 Luna | Muse Spark 1.3 | DeepSeek V4 Flash 0731 |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 83% | 84 | 86 | 82 |
| Context window | 17% | 65 | 65 | 64 |
| Overall | 100% | 81/100 | 82/100 | 79/100 |
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) | 156.5 | 158.2 (best) | 154.4 |
| ECI rank | #38 of 159 | #28 of 159 (best) | #53 of 159 |
| ECI confidence interval | 154.6–158.5 | 156.2–160.5 | 152.6–156.4 |
| ECI evaluation date | Sep 22, 2026 | Sep 2, 2026 | Jul 31, 2026 |
| Documented capabilities | 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. | Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding |
| Model parameters | Not published | Not published | Not published |
| Output modalities | Text | Text | Text |
| GPQA DiamondGraduate-level science questions | 90.5%max | Not published | 91.0%max (best) |
| FrontierMath Tiers 1–3Research-level mathematics | 79.0%max (best) | 74.4%xhigh | 57.5%max |
| OTIS Mock AIME 2024–2025Competition mathematics | 98.9%max | 99.2%xhigh (best) | 94.4%max |
| SimpleQA VerifiedShort factual questions | 41.4%max (best) | Not published | 33.6%max |
| Publisher-reported benchmarks | |||
| DeepSWE 1.1max effort | 66.6scoreReported resultSep 22, 2026 | Not published | Not published |
| GDPval-AA 2max effort | Not published | 1754EloReported resultSep 2, 2026 | Not published |
| JobBenchmax effort | Not published | 64.9mean rubric scoreReported resultSep 2, 2026 | Not published |
| OSWorld 2.0 · 08.08max effort | Not published | 66.9mean partial scoreReported resultSep 2, 2026 | Not published |
| OSWorld 2.0 · 08.08max effort | Not published | 32binary completion rateReported resultSep 2, 2026 | Not published |
| DeepSearchQAmax effort | Not published | 90.3F1Reported resultSep 2, 2026 | Not published |
| Agentic IF Index · Meta internalmax effort | Not published | 57.8scoreReported resultSep 2, 2026 | Not published |
| AutomationBench · public v3 task setmax effort | Not published | 49.6pass@1Reported resultSep 2, 2026 | Not published |
| MRCR 2 · 8-needle, 256K-512Kmax effort | Not published | 98.5mean sequence-match ratioReported resultSep 2, 2026 | Not published |
| MRCR 2 · 8-needle, 512K-1Mmax effort | Not published | 98.1mean sequence-match ratioReported resultSep 2, 2026 | Not published |
| DeepSWE 1.1max effort | Not published | 75.4pass@1Reported resultSep 2, 2026 | Not published |
| SWE-Atlas Codebase QnAmax effort | Not published | 59.4pass@1Reported resultSep 2, 2026 | Not published |
| Terminal-Bench 2.1max effort | Not published | 88.8pass@1Reported resultSep 2, 2026 | Not published |
| Terminal-Bench 2.1max | Not published | Not published | 82.7pass@1Reported result |
| NL2Repomax effort | Not published | Not published | 54.2resolve rateReported resultJul 31, 2026 |
| CyberGymmax effort | Not published | Not published | 76.7scoreReported resultJul 31, 2026 |
| DeepSWEmax effort | Not published | Not published | 54.4resolve rateReported resultJul 31, 2026 |
| Toolathlon-Verifiedmax effort | Not published | Not published | 70.3scoreReported resultJul 31, 2026 |
| Agents' Last Exammax effort | Not published | Not published | 25.2scoreReported resultJul 31, 2026 |
| AutomationBench · publicmax effort | Not published | Not published | 25.1success rateReported resultJul 31, 2026 |
| DSBench-FullStack · internalmax effort | Not published | Not published | 68.7scoreReported resultJul 31, 2026 |
| DSBench-Hard · internalmax effort | Not published | Not published | 59.6scoreReported resultJul 31, 2026 |
| Price per million tokens | |||
| Input | $0.10 (best) | $1.25 | $0.14 |
| Output | $0.50 | $4.25 | $0.30 (best) |
| Cached input | $0.01 (best) | $0.15 | Not published |
| Blended (3:1) | $0.20 | $2.00 | $0.18 (best) |
| Long-context rate | Over 272K: $0.20 / $0.75 | Same rate | Same rate |
| Price source | Official OpenAI API | Official Meta API | Median of 65 providers |
| Limits | |||
| Context window | 1,050,000 tokens (best) | 1,048,576 tokens | 1,000,000 tokens |
| Max input | 922,000 tokens | Not published | Not published |
| Max output | 128,000 tokens | 131,072 tokens | 384,000 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | Yes | No |
| PDFs | Yes | Yes | No |
| Audio | No | Yes | No |
| Video | No | Yes | No |
| Reasoning | Yeslow · medium · high · xhigh · max | Yesminimal · low · medium · high · xhigh · max | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | Yes | Not published |
| File attachments | Yes | Yes | No |
| Temperature control | Yes | Yes | Yes |
| Reasoning control | effort | effort | always |
| Availability | |||
| Weights | Proprietary | Proprietary | OpenMIT |
| API model ID | gpt-6-luna | muse-spark-1.3 | Not published |
| API providers | 49 | 18 | 66 (best) |
| Released | Sep 22, 2026 | Sep 2, 2026 | Jul 31, 2026 |
| Knowledge cutoff | May 18, 2026 | Not published | May 2025 |
| Publisher documentation | Publisher documentation | Publisher documentation | Model weights |
| Source metadata updated | Sep 22, 2026 | Sep 2, 2026 | Jul 31, 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.
GPT-6 Luna$2.00
Muse Spark 1.3$21.00
DeepSeek V4 Flash 0731$2.00
Which should you choose?
Which is better: GPT-6 Luna, Muse Spark 1.3 or DeepSeek V4 Flash 0731?
It is close. Our weighted score puts them within a point (GPT-6 Luna 80/100, DeepSeek V4 Flash 0731 80/100, Muse Spark 1.3 68/100), so choose by what matters most for your work: Muse Spark 1.3 for raw capability and DeepSeek V4 Flash 0731 on price. The score weighs capability 59%, price 29%, context window 12%.It is close. Our weighted score puts them within 1 points (Muse Spark 1.3 82/100, GPT-6 Luna 81/100, DeepSeek V4 Flash 0731 79/100), so choose by what matters most for your work: Muse Spark 1.3 for raw capability and DeepSeek V4 Flash 0731 on price. The score weighs capability 83%, context window 17%.
Which is cheaper, GPT-6 Luna, Muse Spark 1.3 or DeepSeek V4 Flash 0731?
DeepSeek V4 Flash 0731 is cheaper at $0.14 input / $0.30 output per million tokens (median across 65 API providers). GPT-6 Luna costs $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). At a typical mix of three input tokens to one output token, that is $0.18 per million tokens for DeepSeek V4 Flash 0731 versus $0.20 for GPT-6 Luna (1.1× as much) and $2.00 for Muse Spark 1.3 (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), GPT-6 Luna 156.5 (#38 of 159) and DeepSeek V4 Flash 0731 154.4 (#53 of 159). The confidence ranges of the top two overlap (156.2–160.5 vs 154.6–158.5), so treat the gap as small. On individual benchmarks: FrontierMath Tiers 1–3 — GPT-6 Luna 79.0%, Muse Spark 1.3 74.4%, DeepSeek V4 Flash 0731 57.5%; OTIS Mock AIME 2024–2025 — Muse Spark 1.3 99.2%, GPT-6 Luna 98.9%, DeepSeek V4 Flash 0731 94.4%.
Which is better for coding?
There are no published SWE-bench Verified results for GPT-6 Luna, Muse Spark 1.3 and DeepSeek V4 Flash 0731 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 DeepSeek V4 Flash 0731. Maximum output per response: GPT-6 Luna up to 128,000, Muse Spark 1.3 up to 131,072, DeepSeek V4 Flash 0731 up to 384,000 tokens.
Which can read images, PDFs, audio or video?
GPT-6 Luna accepts text, images and PDFs; Muse Spark 1.3 accepts text, images, PDFs, audio and video; DeepSeek V4 Flash 0731 accepts text. Muse Spark 1.3 handles the widest range of inputs.
Are any of these open source?
DeepSeek V4 Flash 0731 publishes its weights (MIT) 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; DeepSeek V4 Flash 0731 came out Jul 31, 2026. Knowledge cutoff: GPT-6 Luna May 18, 2026, DeepSeek V4 Flash 0731 May 2025.
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.