DeepSeek V4 Flash 0731 vs Qwen3.5 Flash vs GPT-6 Luna
GPT-6 Luna comes out ahead, 78 to 69 and 65 on our weighted score, though DeepSeek V4 Flash 0731 is 13% cheaper per token.
DeepSeekDeepSeek V4 Flash 0731
69/100- ECI154.5
- Price$0.14 / $0.28
- Context1M
Alibaba (Qwen)Qwen3.5 Flash
65/100- ECI144.0
- Price$0.10 / $0.40
- Context1M
- Our pick
OpenAIGPT-6 Luna
78/100- ECI—
- Price$0.10 / $0.50
- Context1.05M
GPT-6 Luna is our pick
GPT-6 Luna is the better all-round choice, scoring 78/100 against DeepSeek V4 Flash 0731 (69) and Qwen3.5 Flash (65). It leads on capability. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%. Capability uses the average of shared benchmarks (GPQA Diamond, FrontierMath Tiers 1–3, OTIS Mock AIME 2024–2025 and SimpleQA Verified), because GPT-6 Luna has no Capabilities Index score yet.
- CapabilityGPT-6 LunaShared benchmarks: GPT-6 Luna 77.4% · DeepSeek V4 Flash 0731 69.2% · Qwen3.5 Flash 51.3%
- Lowest priceDeepSeek V4 Flash 0731 and Qwen3.5 FlashDeepSeek V4 Flash 0731 $0.175 · Qwen3.5 Flash $0.175 · GPT-6 Luna $0.20 per 1M tokens (3:1 blend)
- Longest contextGPT-6 LunaGPT-6 Luna 1,050,000 · DeepSeek V4 Flash 0731 1,000,000 · Qwen3.5 Flash 1,000,000 tokens
- Widest inputsQwen3.5 Flash and GPT-6 LunaDeepSeek V4 Flash 0731: Text · Qwen3.5 Flash: Text, Images, Video · GPT-6 Luna: Text, Images, PDFs
- Self-hostingDeepSeek V4 Flash 0731Publishes downloadable weights (MIT)
| Measure | Weight | DeepSeek V4 Flash 0731 | Qwen3.5 Flash | GPT-6 Luna |
|---|---|---|---|---|
| CapabilityShared benchmarks | 50% | 69 | 51 | 77 |
| Price | 25% | 86 | 86 | 83 |
| Inputs & features | 15% | 45 | 80 | 80 |
| Context window | 10% | 60 | 60 | 61 |
| Overall | 100% | 69/100 | 65/100 | 78/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | DeepSeek V4 Flash 0731DeepSeek | Qwen3.5 FlashAlibaba (Qwen) | GPT-6 LunaOpenAI |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 154.5 (best) | 144.0 | — |
| ECI rank | #32 of 148 (best) | #82 of 148 | — |
| GPQA DiamondGraduate-level science questions | 91.0% (best) | 82.3% | 90.5% |
| FrontierMath Tiers 1–3Research-level mathematics | 57.5% | 18.3% | 79.0% (best) |
| OTIS Mock AIME 2024–2025Competition mathematics | 94.4% | 84.4% | 98.9% (best) |
| SimpleQA VerifiedShort factual questions | 33.6% | 20.3% | 41.4% (best) |
| Price per million tokens | |||
| Input | $0.14 | $0.10 (best) | $0.10 (best) |
| Output | $0.28 (best) | $0.40 | $0.50 |
| Cached input | — | $0.01 | $0.01 |
| Blended (3:1) | $0.175 (best) | $0.175 (best) | $0.20 |
| Long-context rate | Same rate | Same rate | Over 272K: $0.20 / $0.75 |
| Price source | Median of 48 providers | Official Alibaba API | Official OpenAI API |
| Limits | |||
| Context window | 1,000,000 tokens | 1,000,000 tokens | 1,050,000 tokens (best) |
| Max output | 384,000 tokens (best) | 65,536 tokens | 128,000 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | Yes | Yes |
| PDFs | No | No | Yes |
| Audio | No | No | No |
| Video | No | Yes | No |
| Reasoning | Yes | Yes | Yeslow · medium · high · xhigh · max |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | Yes | Yes |
| Availability | |||
| Weights | OpenMIT | Proprietary | Proprietary |
| API model ID | — | qwen3.5-flash | gpt-6-luna |
| API providers | 49 (best) | 8 | 24 |
| Released | Jul 31, 2026 | Feb 23, 2026 | Sep 22, 2026 |
| Knowledge cutoff | May 2025 | — | May 18, 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 Flash 0731$1.96
- Qwen3.5 Flash$1.80
- GPT-6 Luna$2.00
Which should you choose?
Which is better: DeepSeek V4 Flash 0731, Qwen3.5 Flash or GPT-6 Luna?
GPT-6 Luna is the better all-round choice, scoring 78/100 against DeepSeek V4 Flash 0731 (69) and Qwen3.5 Flash (65). It leads on capability. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%. Capability uses the average of shared benchmarks (GPQA Diamond, FrontierMath Tiers 1–3, OTIS Mock AIME 2024–2025 and SimpleQA Verified), because GPT-6 Luna has no Capabilities Index score yet.
Which is cheaper, DeepSeek V4 Flash 0731, Qwen3.5 Flash or GPT-6 Luna?
DeepSeek V4 Flash 0731 is cheaper at $0.14 input / $0.28 output per million tokens (median across 48 API providers). Qwen3.5 Flash costs $0.10 input / $0.40 output per million tokens (official Alibaba API price); GPT-6 Luna costs $0.10 input / $0.50 output per million tokens (official OpenAI API price). At a typical mix of three input tokens to one output token, that is $0.175 per million tokens for DeepSeek V4 Flash 0731 versus $0.175 for Qwen3.5 Flash (1× as much) and $0.20 for GPT-6 Luna (1.1× as much).
Which scores higher on benchmarks?
Not every model here has a Capabilities Index score, so we compare the average of shared benchmarks (GPQA Diamond, FrontierMath Tiers 1–3, OTIS Mock AIME 2024–2025 and SimpleQA Verified): GPT-6 Luna 77.4%, DeepSeek V4 Flash 0731 69.2% and Qwen3.5 Flash 51.3%. On individual benchmarks: GPQA Diamond — DeepSeek V4 Flash 0731 91.0%, GPT-6 Luna 90.5%, Qwen3.5 Flash 82.3%; FrontierMath Tiers 1–3 — GPT-6 Luna 79.0%, DeepSeek V4 Flash 0731 57.5%, Qwen3.5 Flash 18.3%; OTIS Mock AIME 2024–2025 — GPT-6 Luna 98.9%, DeepSeek V4 Flash 0731 94.4%, Qwen3.5 Flash 84.4%; SimpleQA Verified — GPT-6 Luna 41.4%, DeepSeek V4 Flash 0731 33.6%, Qwen3.5 Flash 20.3%.
Which is better for coding?
There are no published SWE-bench Verified results for DeepSeek V4 Flash 0731, Qwen3.5 Flash and GPT-6 Luna yet, so there is no like-for-like coding score. On overall capability, GPT-6 Luna 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 has the largest context window at 1,050,000 tokens, against 1,000,000 for DeepSeek V4 Flash 0731 and 1,000,000 for Qwen3.5 Flash. Maximum output per response: DeepSeek V4 Flash 0731 up to 384,000, Qwen3.5 Flash up to 65,536, GPT-6 Luna up to 128,000 tokens.
Which can read images, PDFs, audio or video?
DeepSeek V4 Flash 0731 accepts text; Qwen3.5 Flash accepts text, images and video; GPT-6 Luna accepts text, images and PDFs. Qwen3.5 Flash 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; Qwen3.5 Flash and GPT-6 Luna is proprietary.
Which is newer?
GPT-6 Luna is the newest, released Sep 22, 2026. DeepSeek V4 Flash 0731 came out Jul 31, 2026; Qwen3.5 Flash came out Feb 23, 2026. Knowledge cutoff: DeepSeek V4 Flash 0731 May 2025, GPT-6 Luna May 18, 2026.
How do you decide the winner?
Each model gets a 0–100 score on 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 Oct 4, 2026.
Related comparisons
- DeepSeek V4 Flash 0731 vs GLM-5.3-FlashZ.ai (Zhipu) · ECI 151.9
- DeepSeek V4 Flash 0731 vs Gemini 3.6 FlashGoogle · ECI 154.3
- DeepSeek V4 Flash 0731 vs Inkling SmallThinking Machines · ECI 150.2
- Qwen3.5 Flash vs Gemma 4 31B ITGoogle · ECI 142.8
- Qwen3.5 Flash vs DeepSeek V4 FlashDeepSeek · ECI 146.1
- Qwen3.5 Flash vs GPT-5.4 nanoOpenAI · ECI 145.8
