DeepSeek V4 Flash vs Grok 4.6 vs Qwen3.5 Flash
Qwen3.5 Flash comes out ahead, 75 to 71 and 65 on our weighted score.
DeepSeekDeepSeek V4 Flash
71/100- ECI146.1
- Price$0.14 / $0.28
- Context1M
xAIGrok 4.6
65/100- ECI156.6
- Price$2.00 / $6.00
- Context500K
- Our pick
Alibaba (Qwen)Qwen3.5 Flash
75/100- ECI144.0
- Price$0.10 / $0.40
- Context1M
Qwen3.5 Flash is our pick
Qwen3.5 Flash is the better all-round choice, scoring 75/100 against DeepSeek V4 Flash (71) and Grok 4.6 (65). It leads on inputs & features. Grok 4.6 wins on capability. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
- CapabilityGrok 4.6Capabilities Index (ECI): Grok 4.6 156.6 · DeepSeek V4 Flash 146.1 · Qwen3.5 Flash 144.0
- Lowest priceDeepSeek V4 Flash and Qwen3.5 FlashDeepSeek V4 Flash $0.175 · Qwen3.5 Flash $0.175 · Grok 4.6 $3.00 per 1M tokens (3:1 blend)
- Longest contextDeepSeek V4 Flash and Qwen3.5 FlashDeepSeek V4 Flash 1,000,000 · Qwen3.5 Flash 1,000,000 · Grok 4.6 500,000 tokens
- Widest inputsQwen3.5 FlashDeepSeek V4 Flash: Text · Grok 4.6: Text, Images · Qwen3.5 Flash: Text, Images, Video
- Self-hostingDeepSeek V4 FlashPublishes downloadable weights
| Measure | Weight | DeepSeek V4 Flash | Grok 4.6 | Qwen3.5 Flash |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 50% | 73 | 86 | 71 |
| Price | 25% | 86 | 27 | 86 |
| Inputs & features | 15% | 45 | 70 | 80 |
| Context window | 10% | 60 | 48 | 60 |
| Overall | 100% | 71/100 | 65/100 | 75/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | DeepSeek V4 FlashDeepSeek | Grok 4.6xAI | Qwen3.5 FlashAlibaba (Qwen) |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 146.1 | 156.6 (best) | 144.0 |
| ECI rank | #71 of 148 | #19 of 148 (best) | #82 of 148 |
| GPQA DiamondGraduate-level science questions | — | 94.0% (best) | 82.3% |
| FrontierMath Tiers 1–3Research-level mathematics | — | 66.0% (best) | 18.3% |
| OTIS Mock AIME 2024–2025Competition mathematics | — | 99.2% (best) | 84.4% |
| SimpleQA VerifiedShort factual questions | — | 49.3% (best) | 20.3% |
| Price per million tokens | |||
| Input | $0.14 | $2.00 | $0.10 (best) |
| Output | $0.28 (best) | $6.00 | $0.40 |
| Cached input | — | $0.50 | $0.01 (best) |
| Blended (3:1) | $0.175 (best) | $3.00 | $0.175 (best) |
| Long-context rate | Same rate | Over 200K: $4.00 / $12.00 | Same rate |
| Price source | Median of 42 providers | Official xAI API | Official Alibaba API |
| Limits | |||
| Context window | 1,000,000 tokens (best) | 500,000 tokens | 1,000,000 tokens (best) |
| Max output | 384,000 tokens | 500,000 tokens (best) | 65,536 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | Yes | Yes |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | Yes |
| Reasoning | Yes | Yeslow · medium · high · xhigh | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | Yes | Yes |
| Availability | |||
| Weights | Open | Proprietary | Proprietary |
| API model ID | — | grok-4.6 | qwen3.5-flash |
| API providers | 48 (best) | 29 | 8 |
| Released | Apr 24, 2026 | Aug 12, 2026 | Feb 23, 2026 |
| Knowledge cutoff | May 2025 | Feb 1, 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$1.96
- Grok 4.6$32.00
- Qwen3.5 Flash$1.80
Which should you choose?
Which is better: DeepSeek V4 Flash, Grok 4.6 or Qwen3.5 Flash?
Qwen3.5 Flash is the better all-round choice, scoring 75/100 against DeepSeek V4 Flash (71) and Grok 4.6 (65). It leads on inputs & features. Grok 4.6 wins on capability. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
Which is cheaper, DeepSeek V4 Flash, Grok 4.6 or Qwen3.5 Flash?
DeepSeek V4 Flash is cheaper at $0.14 input / $0.28 output per million tokens (median across 42 API providers). Qwen3.5 Flash costs $0.10 input / $0.40 output per million tokens (official Alibaba API price); Grok 4.6 costs $2.00 input / $6.00 output per million tokens (official xAI 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 versus $0.175 for Qwen3.5 Flash (1× as much) and $3.00 for Grok 4.6 (17× as much).
Which scores higher on benchmarks?
Grok 4.6 scores higher on the Capabilities Index (ECI): Grok 4.6 156.6 (#19 of 148), DeepSeek V4 Flash 146.1 (#71 of 148) and Qwen3.5 Flash 144.0 (#82 of 148). Their confidence ranges do not overlap (154.7–158.9 vs 143.6–147.9), so the gap is a real one.
Which is better for coding?
There are no published SWE-bench Verified results for DeepSeek V4 Flash, Grok 4.6 and Qwen3.5 Flash yet, so there is no like-for-like coding score. On overall capability, Grok 4.6 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?
DeepSeek V4 Flash and Qwen3.5 Flash have the largest context windows (1,000,000 and 1,000,000 tokens), against 500,000 for Grok 4.6. Maximum output per response: DeepSeek V4 Flash up to 384,000, Grok 4.6 up to 500,000, Qwen3.5 Flash up to 65,536 tokens.
Which can read images, PDFs, audio or video?
DeepSeek V4 Flash accepts text; Grok 4.6 accepts text and images; Qwen3.5 Flash accepts text, images and video. Qwen3.5 Flash handles the widest range of inputs.
Are any of these open source?
DeepSeek V4 Flash publishes its weights and can be self-hosted; Grok 4.6 and Qwen3.5 Flash is proprietary.
Which is newer?
Grok 4.6 is the newest, released Aug 12, 2026. DeepSeek V4 Flash came out Apr 24, 2026; Qwen3.5 Flash came out Feb 23, 2026. Knowledge cutoff: DeepSeek V4 Flash May 2025, Grok 4.6 Feb 1, 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.
