DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows. This is the GA release of DeepSeek V4 Flash.
- Organization
- DeepSeek
- Family
- deepseek-flash
- Providers
- 52
- Context
- 1,000,000
- Output limit
- 384,000
- Knowledge
- 2025-05
- Release
- 2026-07-31
- Updated
- —
- Weights
- Open
- Input
- text
- Output
- text
- Capabilities
- —
Providers
Compare every cataloged offer for this model, including its provider, route, token limits, price, and capabilities.
| Provider | Model ID | Context | Output | Input / output · 1M | Reasoning | Tools | Structured | Details |
|---|---|---|---|---|---|---|---|---|
| AIHubMix | deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| AIHubMix | deepseek-v4-flash-0731-fast | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Alibaba | deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Alibaba Token Plan | deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Alibaba Token Plan (China) | deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| AMD | DeepSeek-V4-Flash | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Arcee | deepseek/deepseek-v4-flash-latest | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Baseten | deepseek-ai/DeepSeek-V4-Flash-0731 | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Bothub | deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Boundless | dsv4 | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Charm Hyper | deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Chutes | deepseek-ai/DeepSeek-V4-Flash-0731-TEE | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Cloudflare Workers AI | @cf/deepseek-ai/deepseek-v4-flash-0731 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| CoreWeave | deepseek-ai/DeepSeek-V4-Flash-0731 | 262,144 | 262,144 | — / — | — | — | — | View |
| Cortecs | deepseek-v4-flash-0731 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| Deep Infra | deepseek-ai/DeepSeek-V4-Flash-0731 | 1,048,576 | 384,000 | — / — | — | — | — | View |
| DigitalOcean | deepseek-v4-flash-0731 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| Eden AI | cloudflare/@cf/deepseek-ai/deepseek-v4-flash-0731 | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Eden AI | sference/deepseek-ai/DeepSeek-V4-Flash-0731 | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Eden AI | greenpt/deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Eden AI | flexai/DeepSeek-V4-Flash-0731 | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Eden AI | tensorx/deepseek/deepseek-v4-flash-0731 | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Eden AI | nebius/deepseek-ai/DeepSeek-V4-Flash-0731 | 1,024,000 | 384,000 | — / — | — | — | — | View |
| Eden AI | qwen/deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Eden AI | scaleway/deepseek-v4-flash-0731 | 256,000 | 384,000 | — / — | — | — | — | View |
| Eden AI | flexai/DeepSeek-V4-Flash-0731@eu | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Eden AI | together_ai/deepseek-ai/DeepSeek-V4-Flash-0731 | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Eden AI | databricks/databricks-deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Eden AI | deepinfra/deepseek-ai/DeepSeek-V4-Flash-0731 | 1,048,576 | 384,000 | — / — | — | — | — | View |
| EmpirioLabs AI | deepseek-v4-flash-0731 | 1,000,000 | 393,216 | — / — | — | — | — | View |
| engy | deepseek-v4-flash-0731 | 1,048,576 | 128,000 | — / — | — | — | — | View |
| GreenPT | deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Hugging Face | deepseek-ai/DeepSeek-V4-Flash-0731 | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Inceptron | deepseek-ai/DeepSeek-V4-Flash-0731 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| Kilo Gateway | deepseek/deepseek-v4-flash-0731 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Melious | deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Merge Gateway | deepseek/deepseek-v4-flash-0731 | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Merge Gateway | deepseek/deepseek-v4-flash-0731-fast | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Moark | deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| NanoGPT | deepseek/deepseek-v4-flash-0731:thinking | 1,048,576 | 131,072 | — / — | — | — | — | View |
| NanoGPT | deepseek/deepseek-v4-flash-0731 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Nebius Token Factory | deepseek-ai/DeepSeek-V4-Flash-0731 | 1,024,000 | 1,024,000 | — / — | — | — | — | View |
| Neuralwatt | deepseek-v4-flash-speed | 1,048,560 | 393,216 | — / — | — | — | — | View |
| Novita AI | deepseek/deepseek-v4-flash-0731 | 1,048,576 | 393,216 | — / — | — | — | — | View |
| Ofox | deepseek/deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| OpenCode | deepseek-v4-flash-free | 200,000 | 128,000 | — / — | — | — | — | View |
| OpenCode | deepseek-v4-flash | 1,000,000 | 384,000 | — / — | — | — | — | View |
| OpenReason | deepseek-ai/deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Alibabavia OpenRouter | deepseek/deepseek-v4-flash-0731 unknown | 1,000,000 | 393,216 | $0.374 / $1.122 | Yes | Yes | Yes | View |
| Alibabavia OpenRouter | deepseek/deepseek-v4-flash-0731 unknown | 1,000,000 | 393,216 | $0.352 / $1.056 | Yes | Yes | Yes | View |
| AtlasCloudvia OpenRouter | deepseek/deepseek-v4-flash-0731 fp4 | 1,048,576 | 393,216 | $0.44 / $1.32 | Yes | Yes | Yes | View |
| Baiduvia OpenRouter | deepseek/deepseek-v4-flash-0731 fp8 | 1,048,576 | 131,072 | $0.44 / $1.32 | Yes | Yes | Yes | View |
| BaseTenvia OpenRouter | deepseek/deepseek-v4-flash-0731 fp8 | 1,048,576 | 384,000 | $0.13 / $0.26 | Yes | Yes | No | View |
| Cloudflarevia OpenRouter | deepseek/deepseek-v4-flash-0731 unknown | 1,048,576 | 943,718 | $0.44 / $1.32 | Yes | Yes | Yes | View |
| Coherevia OpenRouter | deepseek/deepseek-v4-flash-0731 unknown | 1,048,576 | 384,000 | $0.14 / $0.28 | Yes | Yes | Yes | View |
| CoreWeavevia OpenRouter | deepseek/deepseek-v4-flash-0731 fp8 | 262,144 | 235,929 | $0.13 / $0.28 | Yes | Yes | No | View |
| DeepInfravia OpenRouter | deepseek/deepseek-v4-flash-0731 fp8 | 1,048,576 | 384,000 | $0.06 / $0.18 | Yes | Yes | Yes | View |
| DigitalOceanvia OpenRouter | deepseek/deepseek-v4-flash-0731 unknown | 1,048,576 | 943,718 | $0.119 / $0.238 | Yes | Yes | Yes | View |
| GMICloudvia OpenRouter | deepseek/deepseek-v4-flash-0731 fp8 | 1,048,575 | 943,717 | $0.286 / $0.858 | Yes | Yes | No | View |
| Inceptronvia OpenRouter | deepseek/deepseek-v4-flash-0731 fp4 | 1,048,576 | 943,718 | $0.1015 / $0.60 | Yes | Yes | Yes | View |
| Mancer 2via OpenRouter | deepseek/deepseek-v4-flash-0731 fp8 | 1,048,576 | 943,718 | $0.20 / $0.60 | Yes | Yes | Yes | View |
| Novitavia OpenRouter | deepseek/deepseek-v4-flash-0731 fp8 | 1,048,576 | 393,216 | $0.4092 / $1.2276 | Yes | Yes | No | View |
| OpenInferencevia OpenRouter | deepseek/deepseek-v4-flash-0731 fp4 | 1,048,576 | 943,718 | $0.01089 / $0.27 | Yes | Yes | Yes | View |
| Parasailvia OpenRouter | deepseek/deepseek-v4-flash-0731 fp8 | 1,048,576 | 943,718 | $0.14 / $0.28 | Yes | Yes | Yes | View |
| Phalavia OpenRouter | deepseek/deepseek-v4-flash-0731 unknown | 1,048,576 | 393,216 | $0.308 / $0.924 | Yes | Yes | Yes | View |
| Rekavia OpenRouter | deepseek/deepseek-v4-flash-0731 unknown | 262,144 | 131,072 | $0.021 / $0.528 | Yes | Yes | Yes | View |
| Relacevia OpenRouter | deepseek/deepseek-v4-flash-0731 fp4 | 1,048,576 | 943,718 | $0.0103 / $1.28 | Yes | Yes | No | View |
| Sail Researchvia OpenRouter | deepseek/deepseek-v4-flash-0731 fp4 | 1,048,576 | 943,718 | $0.019 / $0.42 | Yes | Yes | Yes | View |
| Sail Researchvia OpenRouter | deepseek/deepseek-v4-flash-0731 fp4 | 1,048,576 | 943,718 | $0.10 / $0.30 | Yes | Yes | Yes | View |
| SiliconFlowvia OpenRouter | deepseek/deepseek-v4-flash-0731 fp8 | 1,048,576 | 393,216 | $0.22 / $0.66 | Yes | Yes | Yes | View |
| StreamLakevia OpenRouter | deepseek/deepseek-v4-flash-0731 fp8 | 1,024,000 | 384,000 | $0.088 / $0.264 | Yes | Yes | No | View |
| Togethervia OpenRouter | deepseek/deepseek-v4-flash-0731 unknown | 1,048,576 | 943,718 | $0.14 / $0.28 | Yes | Yes | Yes | View |
| Venicevia OpenRouter | deepseek/deepseek-v4-flash-0731 unknown | 1,000,000 | 32,768 | $0.13 / $0.26 | Yes | Yes | Yes | View |
| Wafervia OpenRouter | deepseek/deepseek-v4-flash-0731 unknown | 1,048,576 | 943,718 | $0.13 / $0.23 | Yes | Yes | Yes | View |
| OrcaRouter | deepseek/deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Perplexity Agent | deepseek/deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Prime Inference | deepseek/deepseek-v4-flash-0731 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Requesty | deepseek-v4-flash-0731 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Requesty | deepseek-v4-flash-0731@eu | 1,048,576 | 131,072 | — / — | — | — | — | View |
| RunInfra | deepseek-ai/DeepSeek-V4-Flash-0731 | 1,048,576 | 32,768 | — / — | — | — | — | View |
| Scaleway | deepseek-v4-flash-0731 | 256,000 | 32,768 | — / — | — | — | — | View |
| SCNet Token Plan | DeepSeek-V4-Flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| SiliconFlow | deepseek-ai/DeepSeek-V4-Flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Standard Compute | deepseek/deepseek-v4-flash-0731 | 1,000,000 | 50,000 | — / — | — | — | — | View |
| TensorX | deepseek/deepseek-v4-flash-0731 | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Together AI | deepseek-ai/DeepSeek-V4-Flash-0731 | 1,048,576 | 384,000 | — / — | — | — | — | View |
| Umans AI | umans-deepseek-v4-flash-0731 | 1,048,576 | 393,215 | — / — | — | — | — | View |
| Umans AI Coding Plan | umans-deepseek-v4-flash-0731 | 1,048,576 | 393,215 | — / — | — | — | — | View |
| Venice AI | deepseek-v4-flash-0731 | 1,000,000 | 32,768 | — / — | — | — | — | View |
| Venice AI | deepseek-v4-flash-0731-fast | 1,000,000 | 32,768 | — / — | — | — | — | View |
| Vercel AI Gateway | deepseek/deepseek-v4-flash-0731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Vivgrid | deepseek-v4-flash | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Volcengine Ark | deepseek-v4-flash-ga-260731 | 1,000,000 | 384,000 | — / — | — | — | — | View |
| Vultr | deepseek-v4-flash-0731 | 1,048,576 | 384,000 | — / — | — | — | — | View |
Prices and limits apply to each offer. A dash means the value is not provided. OpenRouter offers show the hosting provider and routing channel separately.
Pricing
Compare token rates and cache charges across available routes and hosting configurations.
Lowest listed input and output prices across the offers above; they may belong to different providers. These are base rates, before conditional pricing, discounts or additional fees.
deepseek/deepseek-v4-flash-0731Observed - Input
- $0.0099
- Output
- $1.28
- Cache read
- $0.0099
Hosting prices through OpenRouter All endpoints and cache rates
| Endpoint | Input / 1M | Output / 1M | Cache read / 1M | Cache write / 1M |
|---|---|---|---|---|
| Alibabaalibaba/us-east-1 | $0.374 | $1.122 | $0.0374 | — |
| Alibabaalibaba | $0.352 | $1.056 | $0.0352 | — |
| AtlasCloudatlas-cloud/fp4 | $0.44 | $1.32 | $0.028 | — |
| Baidubaidu/fp8 | $0.44 | $1.32 | $0.014 | — |
| BaseTenbaseten/fp8 | $0.13 | $0.26 | $0.028 | — |
| Cloudflarecloudflare | $0.44 | $1.32 | $0.014 | — |
| Coherecohere | $0.14 | $0.28 | $0.07 | — |
| CoreWeavecoreweave/fp8 | $0.13 | $0.28 | $0.07 | — |
| DeepInfradeepinfra/fp8 | $0.06 | $0.18 | $0.015 | — |
| DigitalOceandigitalocean | $0.119 | $0.238 | $0.0238 | — |
| GMICloudgmicloud/fp8 | $0.286 | $0.858 | $0.0091 | — |
| Inceptroninceptron/fp4 | $0.1015 | $0.60 | $0.027 | — |
| Mancer 2mancer/fp8 | $0.20 | $0.60 | — | — |
| Novitanovita/fp8 | $0.4092 | $1.2276 | $0.02604 | — |
| OpenInferenceopen-inference/fp4 | $0.01089 | $0.27 | $0.00792 | — |
| Parasailparasail/fp8 | $0.14 | $0.28 | $0.05 | — |
| Phalaphala | $0.308 | $0.924 | $0.0196 | — |
| Rekareka | $0.021 | $0.528 | $0.0056 | — |
| Relacerelace/fp4 | $0.0103 | $1.28 | $0.0103 | — |
| Sail Researchsail-research/us | $0.019 | $0.42 | $0.012 | — |
| Sail Researchsail-research/fp4 | $0.10 | $0.30 | $0.03 | — |
| SiliconFlowsiliconflow/fp8 | $0.22 | $0.66 | $0.028 | — |
| StreamLakestreamlake/fp8 | $0.088 | $0.264 | $0.0028 | — |
| Togethertogether | $0.14 | $0.28 | $0.03 | — |
| Venicevenice | $0.13 | $0.26 | $0.028 | — |
| Waferwafer/fast | $0.13 | $0.23 | $0.12 | — |
Each row is a hosting configuration. Open an offer in Providers for its full pricing conditions and additional fees. Endpoint observations may differ from the routing catalog's base price.
Performance
Review reported latency and output-throughput percentiles from the latest observed rolling window.
Time to first token · milliseconds
All latency percentiles 25 endpoints
| Endpoint | P50 | P75 | P90 | P99 |
|---|---|---|---|---|
| Alibabaalibaba/us-east-1 | 2,131 ms | 4,470 ms | 5,333.6 ms | 7,110.68 ms |
| Alibabaalibaba | 1,224 ms | 1,586 ms | 1,884 ms | 5,092 ms |
| AtlasCloudatlas-cloud/fp4 | 1,669 ms | 2,045 ms | 3,170.8 ms | 5,815.86 ms |
| Baidubaidu/fp8 | 964 ms | 1,266 ms | 1,696 ms | 11,320.4 ms |
| Cloudflarecloudflare | 3,288.5 ms | 4,183 ms | 5,016.3 ms | 12,183.46 ms |
| Coherecohere | 398 ms | 818 ms | 1,752.7 ms | 10,249.36 ms |
| CoreWeavecoreweave/fp8 | 451 ms | 765.25 ms | 1,497 ms | 5,551.22 ms |
| DeepInfradeepinfra/fp8 | 804.5 ms | 1,358.25 ms | 2,514.8 ms | 7,782.35 ms |
| DigitalOceandigitalocean | 745 ms | 1,079 ms | 1,649 ms | 6,571.27 ms |
| GMICloudgmicloud/fp8 | 1,842 ms | 2,227 ms | 2,838 ms | 6,016.26 ms |
| Inceptroninceptron/fp4 | 837 ms | 1,620 ms | 3,457.6 ms | 38,262.68 ms |
| Mancer 2mancer/fp8 | 1,046.5 ms | 1,668 ms | 2,621.2 ms | 6,660.53 ms |
| Novitanovita/fp8 | 2,067 ms | 2,564.75 ms | 3,370.3 ms | 11,237.36 ms |
| OpenInferenceopen-inference/fp4 | 1,574.5 ms | 2,566 ms | 4,056 ms | 9,361.54 ms |
| Parasailparasail/fp8 | 732 ms | 1,012 ms | 1,954 ms | 15,893.29 ms |
| Phalaphala | 4,010.5 ms | 5,803 ms | 8,769.1 ms | 46,427.22 ms |
| Rekareka | 594 ms | 890.25 ms | 2,612.6 ms | 20,989.52 ms |
| Relacerelace/fp4 | 594 ms | 1,009 ms | 1,838.9 ms | 8,630.43 ms |
| Sail Researchsail-research/us | 2,036 ms | 2,668.25 ms | 3,343.5 ms | 7,744.63 ms |
| Sail Researchsail-research/fp4 | 2,221 ms | 2,822 ms | 3,617.9 ms | 7,263.95 ms |
| SiliconFlowsiliconflow/fp8 | 1,300 ms | 1,567 ms | 2,025.8 ms | 5,020.48 ms |
| StreamLakestreamlake/fp8 | 1,233 ms | 1,931 ms | 2,918 ms | 8,598.5 ms |
| Togethertogether | 971.5 ms | 2,895 ms | 6,348.9 ms | 31,151.3 ms |
| Venicevenice | 1,202 ms | 1,847.5 ms | 3,237 ms | 50,473 ms |
| Waferwafer/fast | 913 ms | 1,556 ms | 2,356.9 ms | 7,567.97 ms |
Output throughput · tokens per second
All throughput percentiles 25 endpoints
| Endpoint | P50 | P75 | P90 | P99 |
|---|---|---|---|---|
| Alibabaalibaba/us-east-1 | 74.5 t/s | 112.25 t/s | 139.1 t/s | 186.23 t/s |
| Alibabaalibaba | 120 t/s | 149 t/s | 167 t/s | 195 t/s |
| AtlasCloudatlas-cloud/fp4 | 51 t/s | 69 t/s | 90 t/s | 141.11 t/s |
| Baidubaidu/fp8 | 62 t/s | 84.25 t/s | 115 t/s | 179.77 t/s |
| Cloudflarecloudflare | 40 t/s | 51 t/s | 66 t/s | 77 t/s |
| Coherecohere | 125 t/s | 162 t/s | 207 t/s | 337 t/s |
| CoreWeavecoreweave/fp8 | 114 t/s | 141 t/s | 173 t/s | 230.09 t/s |
| DeepInfradeepinfra/fp8 | 37 t/s | 49 t/s | 60 t/s | 86 t/s |
| DigitalOceandigitalocean | 45 t/s | 49 t/s | 52 t/s | 57.09 t/s |
| GMICloudgmicloud/fp8 | 63 t/s | 77 t/s | 100 t/s | 161 t/s |
| Inceptroninceptron/fp4 | 55 t/s | 90 t/s | 126 t/s | 211 t/s |
| Mancer 2mancer/fp8 | 44 t/s | 54 t/s | 65.4 t/s | 89.56 t/s |
| Novitanovita/fp8 | 69 t/s | 80 t/s | 88 t/s | 165.14 t/s |
| OpenInferenceopen-inference/fp4 | 20 t/s | 23 t/s | 26 t/s | 29 t/s |
| Parasailparasail/fp8 | 82.5 t/s | 129 t/s | 175 t/s | 277.09 t/s |
| Phalaphala | 53 t/s | 74.25 t/s | 91.9 t/s | 140 t/s |
| Rekareka | 143 t/s | 208 t/s | 270 t/s | 389.09 t/s |
| Relacerelace/fp4 | 57 t/s | 103 t/s | 157 t/s | 261 t/s |
| Sail Researchsail-research/us | 38 t/s | 45 t/s | 54 t/s | 78 t/s |
| Sail Researchsail-research/fp4 | 38 t/s | 44 t/s | 51 t/s | 71 t/s |
| SiliconFlowsiliconflow/fp8 | 29 t/s | 33 t/s | 57 t/s | 103.67 t/s |
| StreamLakestreamlake/fp8 | 46 t/s | 62 t/s | 74 t/s | 94 t/s |
| Togethertogether | 55 t/s | 79 t/s | 107 t/s | 172 t/s |
| Venicevenice | 38 t/s | 47 t/s | 59 t/s | 84.2 t/s |
| Waferwafer/fast | 87 t/s | 107 t/s | 132 t/s | 211.09 t/s |
Percentiles describe the observed request distribution. These are snapshots of a rolling window, not a historical time series.
Uptime
Compare successful request rates across the latest 5-minute, 30-minute, and 24-hour observations.
Reported by OpenRouter at the time of collection; rate-limited requests are excluded. A dash means no measurement was reported. This is not a live availability check.
Benchmarks
Review attributed evaluation results on each source's original scale.
deepseek/deepseek-v4-flash-0731Observed | Design Arena | Category | Elo | Win rate | Rank |
|---|---|---|---|---|
| models | 3d | 1,214 | 48.6% | 40 |
| models | asciiart | 1,097 | 33.3% | 61 |
| models | codecategories | 1,233 | 46.4% | 44 |
| models | dataviz | 1,184 | 41.1% | 59 |
| models | gamedev | 1,216 | 44.6% | 46 |
| models | svg | 1,186 | 45.1% | 30 |
| models | uicomponent | 1,239 | 46.8% | 42 |
| models | website | 1,242 | 46.7% | 44 |
Sources: Artificial Analysis and Design Arena, via OpenRouter. Scores retain their original scales and are not OpenAI Suite ratings.
Detailed evaluations
| Evaluation | Configuration | Source | Results | Details |
|---|---|---|---|---|
| search browsecomp | Deepseek V4 Flash 0731deepseek/deepseek-v4-flash-0731 | OpenRouter | 77% | Details |
| search widesearch | Deepseek V4 Flash 0731deepseek/deepseek-v4-flash-0731 | OpenRouter | 77.129% | Details |
| search dsqa | Deepseek V4 Flash 0731deepseek/deepseek-v4-flash-0731 | OpenRouter | 81.655% | Details |
| models · svg | DeepSeek-V4-Flash-0731deepseek/deepseek-v4-flash-20260731 | Design Arena | 1,185 Elo | Details |
| models · dataviz | DeepSeek-V4-Flash-0731deepseek/deepseek-v4-flash-20260731 | Design Arena | 1,180 Elo | Details |
| tau bench verified airline | DeepSeek: DeepSeek V4 Flash 0731deepseek/deepseek-v4-flash-20260731 | OpenRouter | 73.933% | Details |
| models · 3d | DeepSeek-V4-Flash-0731deepseek/deepseek-v4-flash-20260731 | Design Arena | 1,209 Elo | Details |
| models · uicomponent | DeepSeek-V4-Flash-0731deepseek/deepseek-v4-flash-20260731 | Design Arena | 1,233 Elo | Details |
| gpqa diamond | DeepSeek: DeepSeek V4 Flash 0731deepseek/deepseek-v4-flash-20260731 | OpenRouter | 86.549% | Details |
| models · asciiart | DeepSeek-V4-Flash-0731deepseek/deepseek-v4-flash-20260731 | Design Arena | 1,090 Elo | Details |
| Composite indices | DeepSeek V4 Flash 0731 (Max)deepseek/deepseek-v4-flash-20260731 | Artificial Analysis | Intelligence 34.3 · Coding 69.1 · Agentic 41 | Details |
| models · gamedev | DeepSeek-V4-Flash-0731deepseek/deepseek-v4-flash-20260731 | Design Arena | 1,208 Elo | Details |
| models · website | DeepSeek-V4-Flash-0731deepseek/deepseek-v4-flash-20260731 | Design Arena | 1,238 Elo | Details |
| models · codecategories | DeepSeek-V4-Flash-0731deepseek/deepseek-v4-flash-20260731 | Design Arena | 1,230 Elo | Details |
Showing 11 of 14 records in this dataset.
Sources: Artificial Analysis, Design Arena and OpenRouter. Evaluation configurations and scales differ; results are not interchangeable. Updated Oct 24, 2026.
Apps & session costs
Compare observed median session costs for applications using this model, grouped by conversation length.
| Application | Model route | Turns | Median cost / session | Details |
|---|---|---|---|---|
| Kilo Code | deepseek/deepseek-v4-flash-20260731 | 50+ turns | $0.263968 | Details |
| Hermes Agent | deepseek/deepseek-v4-flash-20260731 | 2–9 turns | $0.003259 | Details |
| Codex | deepseek/deepseek-v4-flash-20260731 | 1 turn | $0.000423 | Details |
| Claude Code | deepseek/deepseek-v4-flash-20260731 | 10–49 turns | $0.032943 | Details |
| Hermes Agent | deepseek/deepseek-v4-flash-20260731 | 50+ turns | $0.191248 | Details |
| Claude Code | deepseek/deepseek-v4-flash-20260731 | 50+ turns | $0.188371 | Details |
| Kilo Code | deepseek/deepseek-v4-flash-20260731 | 2–9 turns | $0.006245 | Details |
| Hermes Agent | deepseek/deepseek-v4-flash-20260731 | 1 turn | $0.000348 | Details |
| Claude Code | deepseek/deepseek-v4-flash-20260731 | 1 turn | $0.000757 | Details |
| Kilo Code | deepseek/deepseek-v4-flash-20260731 | 10–49 turns | $0.038249 | Details |
| Hermes Agent | deepseek/deepseek-v4-flash-20260731 | 10–49 turns | $0.027588 | Details |
| Claude Code | deepseek/deepseek-v4-flash-20260731 | 2–9 turns | $0.004451 | Details |
| Kilo Code | deepseek/deepseek-v4-flash-20260731 | 1 turn | $0.000215 | Details |
Source: OpenRouter, as of Oct 20, 2026. Published under CC BY 4.0. These are observed costs per session, not token prices or a forecast for your workload. Apps are compared separately.
Activity
Daily prompt and completion tokens reported for this model's routes in OpenRouter's top 50. Missing days have no published value.
117,062,075,349 tokens
492,645,891,678 tokens
706,786,292,625 tokens
1,015,839,531,451 tokens
1,120,041,012,428 tokens
1,258,592,247,719 tokens
1,440,365,965,940 tokens
1,425,197,504,878 tokens
1,231,278,479,102 tokens
1,334,761,681,026 tokens
1,576,754,642,264 tokens
1,763,515,344,089 tokens
1,724,399,168,396 tokens
1,634,182,107,125 tokens
1,618,292,508,126 tokens
1,407,526,668,596 tokens
1,455,626,623,174 tokens
1,691,677,097,991 tokens
1,741,314,505,634 tokens
1,684,920,552,235 tokens
1,825,475,696,417 tokens
1,849,078,827,192 tokens
1,393,318,211,490 tokens
1,390,118,528,351 tokens
1,710,948,125,723 tokens
1,770,607,978,972 tokens
2,406,082,692,533 tokens
2,001,383,413,526 tokens
1,605,564,470,106 tokens
1,372,084,924,913 tokens
1,439,645,900,160 tokens
1,591,945,119,671 tokens
1,633,618,836,237 tokens
1,728,912,524,247 tokens
1,913,443,637,145 tokens
2,238,811,523,617 tokens
1,732,594,234,286 tokens
1,520,532,796,335 tokens
1,576,698,141,626 tokens
1,721,613,635,988 tokens
1,618,860,272,576 tokens
1,949,990,851,916 tokens
1,909,849,336,222 tokens
1,344,100,323,527 tokens
1,441,764,456,890 tokens
1,737,561,958,149 tokens
1,506,862,691,028 tokens
1,349,217,150,952 tokens
1,327,505,194,194 tokens
1,190,061,791,485 tokens
1,213,436,800,901 tokens
1,115,867,963,555 tokens
1,206,123,550,663 tokens
1,236,716,483,499 tokens
1,200,912,439,316 tokens
1,138,661,880,704 tokens
1,114,855,555,373 tokens
1,085,804,508,442 tokens
841,010,990,367 tokens
948,299,816,031 tokens
945,806,096,452 tokens
968,247,430,381 tokens
922,138,361,503 tokens
748,543,750,000 tokens
708,736,901,475 tokens
748,991,224,099 tokens
853,299,292,680 tokens
837,988,232,787 tokens
713,532,318,593 tokens
668,135,877,749 tokens
680,702,210,115 tokens
542,347,269,702 tokens
620,127,280,606 tokens
601,059,076,490 tokens
598,509,158,568 tokens
995,780,142,653 tokens
691,049,828,563 tokens
620,600,893,373 tokens
584,493,642,131 tokens
608,182,877,846 tokens
626,432,649,405 tokens
585,593,798,357 tokens
575,674,647,614 tokens
586,371,963,995 tokens
601,089,481,070 tokens
Daily observations · 85 days
| Date (UTC) | Reported tokens | Reported routes |
|---|---|---|
| Oct 23, 2026 | 601,089,481,070 | 1 |
| Oct 22, 2026 | 586,371,963,995 | 1 |
| Oct 21, 2026 | 575,674,647,614 | 1 |
| Oct 20, 2026 | 585,593,798,357 | 1 |
| Oct 19, 2026 | 626,432,649,405 | 1 |
| Oct 18, 2026 | 608,182,877,846 | 1 |
| Oct 17, 2026 | 584,493,642,131 | 1 |
| Oct 16, 2026 | 620,600,893,373 | 1 |
| Oct 15, 2026 | 691,049,828,563 | 1 |
| Oct 14, 2026 | 995,780,142,653 | 1 |
| Oct 13, 2026 | 598,509,158,568 | 1 |
| Oct 12, 2026 | 601,059,076,490 | 1 |
| Oct 11, 2026 | 620,127,280,606 | 1 |
| Oct 10, 2026 | 542,347,269,702 | 1 |
| Oct 9, 2026 | 680,702,210,115 | 1 |
| Oct 8, 2026 | 668,135,877,749 | 1 |
| Oct 7, 2026 | 713,532,318,593 | 1 |
| Oct 6, 2026 | 837,988,232,787 | 1 |
| Oct 5, 2026 | 853,299,292,680 | 1 |
| Oct 4, 2026 | 748,991,224,099 | 1 |
| Oct 3, 2026 | 708,736,901,475 | 1 |
| Oct 2, 2026 | 748,543,750,000 | 1 |
| Oct 1, 2026 | 922,138,361,503 | 1 |
| Sep 30, 2026 | 968,247,430,381 | 1 |
| Sep 29, 2026 | 945,806,096,452 | 1 |
| Sep 28, 2026 | 948,299,816,031 | 1 |
| Sep 27, 2026 | 841,010,990,367 | 1 |
| Sep 26, 2026 | 1,085,804,508,442 | 1 |
| Sep 25, 2026 | 1,114,855,555,373 | 1 |
| Sep 24, 2026 | 1,138,661,880,704 | 1 |
| Sep 23, 2026 | 1,200,912,439,316 | 1 |
| Sep 22, 2026 | 1,236,716,483,499 | 1 |
| Sep 21, 2026 | 1,206,123,550,663 | 1 |
| Sep 20, 2026 | 1,115,867,963,555 | 1 |
| Sep 19, 2026 | 1,213,436,800,901 | 1 |
| Sep 18, 2026 | 1,190,061,791,485 | 1 |
| Sep 17, 2026 | 1,327,505,194,194 | 1 |
| Sep 16, 2026 | 1,349,217,150,952 | 1 |
| Sep 15, 2026 | 1,506,862,691,028 | 1 |
| Sep 14, 2026 | 1,737,561,958,149 | 1 |
| Sep 13, 2026 | 1,441,764,456,890 | 1 |
| Sep 12, 2026 | 1,344,100,323,527 | 1 |
| Sep 11, 2026 | 1,909,849,336,222 | 1 |
| Sep 10, 2026 | 1,949,990,851,916 | 1 |
| Sep 9, 2026 | 1,618,860,272,576 | 1 |
| Sep 8, 2026 | 1,721,613,635,988 | 1 |
| Sep 7, 2026 | 1,576,698,141,626 | 1 |
| Sep 6, 2026 | 1,520,532,796,335 | 1 |
| Sep 5, 2026 | 1,732,594,234,286 | 1 |
| Sep 4, 2026 | 2,238,811,523,617 | 1 |
| Sep 3, 2026 | 1,913,443,637,145 | 1 |
| Sep 2, 2026 | 1,728,912,524,247 | 1 |
| Sep 1, 2026 | 1,633,618,836,237 | 1 |
| Aug 31, 2026 | 1,591,945,119,671 | 1 |
| Aug 30, 2026 | 1,439,645,900,160 | 1 |
| Aug 29, 2026 | 1,372,084,924,913 | 1 |
| Aug 28, 2026 | 1,605,564,470,106 | 1 |
| Aug 27, 2026 | 2,001,383,413,526 | 1 |
| Aug 26, 2026 | 2,406,082,692,533 | 1 |
| Aug 25, 2026 | 1,770,607,978,972 | 1 |
| Aug 24, 2026 | 1,710,948,125,723 | 1 |
| Aug 23, 2026 | 1,390,118,528,351 | 1 |
| Aug 22, 2026 | 1,393,318,211,490 | 1 |
| Aug 21, 2026 | 1,849,078,827,192 | 1 |
| Aug 20, 2026 | 1,825,475,696,417 | 1 |
| Aug 19, 2026 | 1,684,920,552,235 | 1 |
| Aug 18, 2026 | 1,741,314,505,634 | 1 |
| Aug 17, 2026 | 1,691,677,097,991 | 1 |
| Aug 16, 2026 | 1,455,626,623,174 | 1 |
| Aug 15, 2026 | 1,407,526,668,596 | 1 |
| Aug 14, 2026 | 1,618,292,508,126 | 1 |
| Aug 13, 2026 | 1,634,182,107,125 | 1 |
| Aug 12, 2026 | 1,724,399,168,396 | 1 |
| Aug 11, 2026 | 1,763,515,344,089 | 1 |
| Aug 10, 2026 | 1,576,754,642,264 | 1 |
| Aug 9, 2026 | 1,334,761,681,026 | 1 |
| Aug 8, 2026 | 1,231,278,479,102 | 1 |
| Aug 7, 2026 | 1,425,197,504,878 | 1 |
| Aug 6, 2026 | 1,440,365,965,940 | 1 |
| Aug 5, 2026 | 1,258,592,247,719 | 1 |
| Aug 4, 2026 | 1,120,041,012,428 | 1 |
| Aug 3, 2026 | 1,015,839,531,451 | 1 |
| Aug 2, 2026 | 706,786,292,625 | 1 |
| Aug 1, 2026 | 492,645,891,678 | 1 |
| Jul 31, 2026 | 117,062,075,349 | 1 |
Source: OpenRouter (openrouter.ai/rankings), as of Oct 24, 2026. CC BY 4.0. Totals include only individually published routes. Tokenizers differ by provider; the aggregated “other” category is never assigned to a model.
Technical details & API
Explore specifications, supported parameters, reasoning controls, and API identifiers for each OpenRouter route.
deepseek/deepseek-v4-flash-0731Observed - Context
- 1,048,576
- Maximum output
- 943,718
- Tokenizer
- DeepSeek
- Input
- text
- Output
- text
- Moderation
- No
- Added to OpenRouter
- 2026-07-31
https://openrouter.ai/api/v1deepseek/deepseek-v4-flash-0731deepseek/deepseek-v4-flash-20260731deepseek-ai/DeepSeek-V4-Flash-0731Reasoning
- Required
- No
- Enabled by default
- Yes
- Default effort
- high
- Supported efforts
- max, high, low
Supported parameters
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_pparallel_tool_callspresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_atop_ktop_logprobstop_p
Frequently asked questions
Answers based on the model metadata and serving offers currently in the catalog.
How much does DeepSeek V4 Flash 0731 cost?
Listed rates start at $0.0103 per million input tokens and $0.18 per million output tokens. Prices vary by provider and configuration; see Pricing for details.
Which providers offer DeepSeek V4 Flash 0731?
There are 94 cataloged offers across 52 providers. Compare model IDs, prices and limits in Providers.
What is the context window?
The cataloged model context is 1,000,000 tokens. Each serving endpoint may apply a different limit.
Does it support tools and structured output?
Tool calling: Not reported. Structured output: Not reported. Support can vary by endpoint.
Resources
MITCatalog history · 3 recorded revisions
Metadata observations since this model was first cataloged. These are not model release versions.
- ·
94e2805e2b72 - ·
39ed45961fb5 - ·
5420dba9e29f