
DeepSeek: DeepSeek V4.1 Flash 1017
deepseek/deepseek-v4.1-flash-1017
Model weightsDeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the deepseek-flash API model name. V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on output from a 552B-parameter backbone, an asymmetric split that keeps per-token compute low relative to the model's total size. Image understanding is native to the architecture, with visual and text embeddings trained jointly from the start of pre-training rather than added afterward as in the earlier experimental V4 Flash Vision ExpOpens in new tab.
It is suited for coding, terminal, and computer-use agents, along with long-horizon tasks that must run to completion across many steps and long-context analysis. Compressed KV caching cuts cache memory to roughly a quarter of the previous Flash generation, significantly reducing costs on agentic workloads. DeepSeek positions it as the cost-efficient tier of the V4.1 family and reports that it exceeds V4 ProOpens in new tab on performance, speed, and task completion time.
Providers
Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).
| $0.12000 | $0.5000 | $0.003000 | 0.76s | 148 tok/s | ||
| $0.15000 | $0.5000 | $0.003000 | 0.94s | 138 tok/s | ||
| $0.08640 | $0.3456 | $0.001728 | 1.03s | 123 tok/s | ||
| $0.18000 | $0.7200 | $0.006000 | 0.79s | 167 tok/s | ||
| $0.07300 | $0.6490 | $0.006000 | 0.47s | 130 tok/s | ||
| $0.03600 | $0.5000 | $0.010000 | 0.36s | 142 tok/s | ||
| $0.07000 | $0.5000 | $0.006900 | 0.38s | 144 tok/s | ||
| $0.04368 | $0.7000 | $0.002400 | 0.98s | 115 tok/s | ||
| $0.20$0.09800 | $0.60$0.2940 | $0.006$0.002940 | 0.78s | 101 tok/s | ||
| $0.30000 | $1.2000 | $0.006000 | 0.61s | 185 tok/s | ||
| $0.30$0.14700 | $0.90$0.4410 | $0.009$0.004410 | 0.97s | 111 tok/s | ||
| $0.07000 | $1.2000 | $0.006000 | 0.99s | 140 tok/s | ||
| $0.30000 | $1.2000 | $0.006000 | 0.52s | 190 tok/s | ||
| $0.25000 | $1.0000 | $0.005000 | 0.79s | 134 tok/s | ||
| $0.29000 | $1.2000 | $0.007000 | 0.77s | 176 tok/s | ||
| $0.08000 | $0.6050 | $0.008000 | 0.25s | 90 tok/s | ||
| $0.30000 | $1.2000 | $0.007000 | 1.58s | 162 tok/s | ||
| $0.21000 | $0.8400 | $0.004200 | 2.14s | 115 tok/s | ||
| $0.08000 | $0.6050 | $0.008000 | 0.39s | 87 tok/s | ||
| $0.15000 | $1.2000 | $0.006000 | 0.93s | 137 tok/s | ||
45% off | $0.30$0.16500 | $1.20$0.6600 | $0.006$0.003300 | 1.25s | 107 tok/s | |
| $0.21000 | $0.6200 | $0.010500 | 1.26s | 117 tok/s | ||
| $0.16000 | $1.1000 | $0.008000 | 1.24s | 103 tok/s | ||
40% off | $0.30$0.18000 | $1.20$0.7200 | $0.006$0.003600 | 1.19s | 123 tok/s | |
| $0.10000 | $0.6000 | $0.003000 | 2.08s | 92 tok/s | ||
| $0.45000 | $1.8000 | $0.009000 | 0.79s | 178 tok/s | ||
| $0.30000 | $1.2000 | $0.006000 | 0.27s | 102 tok/s | ||
| $0.06000 | $0.5000 | $0.030000 | 0.94s | 80 tok/s | ||
| $0.30000 | $1.2000 | $0.006000 | 1.10s | 160 tok/s | ||
63% off | $0.30$0.11100 | $1.20$0.4440 | $0.006$0.002220 | 1.78s | 53 tok/s | |
| $0.30000 | $1.2000 | $0.007000 | 0.63s | 134 tok/s | ||
| $0.28000 | $1.1300 | $0.006000 | 1.32s | 100 tok/s | ||
| $0.30000 | $1.2000 | $0.007500 | 1.16s | 100 tok/s | ||
| $0.30000 | $1.2000 | $0.006000 | 1.77s | 83 tok/s | ||
| $0.30000 | $0.5500 | $0.006000 | 0.80s | 60 tok/s | ||
| $0.30000 | $1.2000 | $0.006000 | 0.71s | 40 tok/s | ||
65% off | $0.30$0.10500 | $1.20$0.4200 | $0.006$0.002100 | 1.29s | 156 tok/s | |
| $0.18000 | $0.6800 | $0.006000 | 1.53s | 147 tok/s | ||
| $0.19500 | $0.7800 | $0.003900 | 2.78s | 89 tok/s | ||
| $0.30000 | $1.2000 | $0.006000 | 1.24s | 120 tok/s | ||
| $0.24000 | $0.9600 | $0.004800 | 2.01s | 94 tok/s | ||
Fast | $0.60000 | $2.4000 | $0.014000 | 0.31s | 459 tok/s | |
| $0.21000 | $0.8400 | $0.004200 | 2.49s | 94 tok/s | ||
| $0.05000 | $0.1000 | $0.007000 | -- | -- | ||
| $0.07000 | $0.1000 | $0.007000 | -- | -- | ||
Not used in Standard routing:Why these endpoints are not used | ||||||
| $0.03000 | $0.2500 | $0.015000 | 0.60s | 87 tok/s | ||
| $0.16$0.07840 | $0.48$0.2352 | $0.0048$0.002352 | 2.48s | 81 tok/s | ||
- Input $0.12 / Output $0.50148 tok/s99.97% uptime
Input $0.15 / Output $0.50138 tok/s100.00% uptime- Input $0.0864 / Output $0.3456123 tok/s100.00% uptime
- Input $0.18 / Output $0.72167 tok/s100.00% uptime
- Input $0.073 / Output $0.649130 tok/s100.00% uptime
- Input $0.036 / Output $0.50142 tok/s100.00% uptime
- Input $0.07 / Output $0.50144 tok/s100.00% uptime
- Input $0.04368 / Output $0.70115 tok/s100.00% uptime
Input $0.098 / Output $0.29451% off101 tok/s100.00% uptime- Input $0.30 / Output $1.20185 tok/s100.00% uptime
Input $0.147 / Output $0.441111 tok/s100.00% uptimePriority+11 more label- Input $0.07 / Output $1.20140 tok/s100.00% uptime
- Input $0.30 / Output $1.20190 tok/s94.89% uptime
- Input $0.25 / Output $1.00134 tok/s100.00% uptime
- Input $0.29 / Output $1.20176 tok/s100.00% uptime
- Input $0.08 / Output $0.60590 tok/s100.00% uptime
Input $0.30 / Output $1.20162 tok/s100.00% uptime- Input $0.21 / Output $0.84115 tok/s99.85% uptime
- Input $0.08 / Output $0.60587 tok/s99.97% uptime
- Input $0.15 / Output $1.20137 tok/s100.00% uptime
- Input $0.165 / Output $0.6645% off107 tok/s99.91% uptime
- Input $0.21 / Output $0.62117 tok/s99.79% uptime
- Input $0.16 / Output $1.10103 tok/s99.82% uptime
- Input $0.18 / Output $0.7240% off123 tok/s99.88% uptime
- Input $0.10 / Output $0.6092 tok/s100.00% uptime
Input $0.45 / Output $1.80178 tok/s100.00% uptime- Input $0.30 / Output $1.20102 tok/s99.94% uptime
- Input $0.06 / Output $0.5080 tok/s99.88% uptime
Input $0.30 / Output $1.20160 tok/s99.97% uptime- Input $0.111 / Output $0.44463% off53 tok/s100.00% uptime
- Input $0.30 / Output $1.20134 tok/s100.00% uptime
- Input $0.28 / Output $1.13100 tok/s100.00% uptime
- Input $0.30 / Output $1.20100 tok/s100.00% uptime
- Input $0.30 / Output $1.2083 tok/s100.00% uptime
- Input $0.30 / Output $0.5560 tok/s100.00% uptime
Input $0.30 / Output $1.2040 tok/s100.00% uptime- Input $0.105 / Output $0.4265% off156 tok/s100.00% uptime
- Input $0.18 / Output $0.68147 tok/s99.90% uptime
- Input $0.195 / Output $0.7889 tok/s100.00% uptime
Input $0.30 / Output $1.20120 tok/s100.00% uptime
Input $0.24 / Output $0.9694 tok/s98.98% uptime- Input $0.60 / Output $2.40459 tok/s100.00% uptimeFast
Input $0.21 / Output $0.8494 tok/s97.48% uptime- Input $0.05 / Output $0.10-- tok/s98.95% uptime
- Input $0.07 / Output $0.10-- tok/s99.32% uptime
- Not used in Standard routing:Why these endpoints are not used
- Input $0.03 / Output $0.2587 tok/s100.00% uptimeFlex
Input $0.0784 / Output $0.235281 tok/s78.27% uptimeFlex+11 more label
Pricing
The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.
Weighted average
DeepSeek V4.1 Flash 1017 — Price history
| $0.00476 | $0.6995 | 94.3% | 3.3% | ||
| $0.00762 | $0.3451 | 93.0% | 1.0% | ||
| $0.00801 | $0.4195 | 94.3% | 0.9% | ||
| $0.00838 | $1.1409 | 96.3% | 3.7% | ||
| $0.00847 | $0.3361 | 95.8% | 11.6% | ||
| $0.00891 | $0.6626 | 96.6% | 1.6% | ||
| $0.00952 | $0.8398 | 97.4% | 4.1% | ||
| $0.00963 | $0.4998 | 94.3% | 4.6% | ||
| $0.01112 | $0.4417 | 95.3% | 1.7% | ||
| $0.01115 | $0.5996 | 91.6% | 0.4% | ||
| $0.01124 | $0.9707 | 97.5% | 22.2% | ||
| $0.01165 | $0.4998 | 91.4% | 2.6% | ||
| $0.01229 | $0.4997 | 91.2% | 6.8% | ||
| $0.01244 | $0.7071 | 92.7% | 3.3% | ||
| $0.01247 | $0.4998 | 93.6% | 2.7% | ||
| $0.01470 | $0.6042 | 90.7% | 5.5% | ||
| $0.01497 | $0.6044 | 90.3% | 0.7% | ||
| $0.01546 | $0.7795 | 93.9% | 0.0% | ||
| $0.01641 | $0.7195 | 94.0% | 3.4% | ||
| $0.01688 | $0.4428 | 86.5% | 0.1% | ||
| $0.01747 | $1.0996 | 93.8% | 0.1% | ||
| $0.02003 | $0.6796 | 91.9% | 0.1% | ||
| $0.02082 | $1.1996 | 95.0% | 6.7% | ||
| $0.02346 | $1.0000 | 92.5% | 0.7% | ||
| $0.02360 | $1.1996 | 94.3% | 0.9% | ||
| $0.02645 | $0.7196 | 87.1% | 0.2% | ||
| $0.02852 | $1.1996 | 92.4% | 1.2% | ||
| $0.03141 | $1.1971 | 91.3% | 0.3% | ||
| $0.03155 | $1.1994 | 91.8% | 0.0% | ||
| $0.03171 | $0.8397 | 86.6% | 0.3% | ||
| $0.03206 | $0.6196 | 89.2% | 0.8% | ||
| $0.03207 | $1.1996 | 81.9% | 0.3% | ||
| $0.03302 | $1.1996 | 90.8% | 1.3% | ||
| $0.03547 | $0.4996 | 59.2% | 0.2% | ||
| $0.03787 | $1.1997 | 89.2% | 0.3% | ||
| $0.04296 | $1.1996 | 87.7% | 0.3% | ||
| $0.04848 | $1.7992 | 91.0% | 0.3% | ||
| $0.04975 | $0.9586 | 80.9% | 0.0% | ||
| $0.06864 | $1.1291 | 77.1% | 0.0% | ||
| $0.07900 | $1.1995 | 75.2% | 0.0% | ||
| $0.08317 | $0.5495 | 73.7% | 0.0% | ||
Fast | $0.10301 | $2.3995 | 84.8% | 0.3% | |
| $0.18910 | $1.1997 | 37.7% | 0.1% | ||
| $0.01086 | $0.1003 | 90.0% | 1.1% | ||
| $0.01088 | $0.0996 | 93.8% | 0.8% | ||
Not used in Standard routing:Why these endpoints are not used | |||||
| $0.00453 | $0.2365 | 97.2% | 1.0% | ||
| $0.01340 | $0.2494 | 72.0% | 2.3% | ||
- Effective in $0.00476 / Effective out $0.6995Token share3.3%
- Effective in $0.00762 / Effective out $0.3451Token share1.0%
- Effective in $0.00801 / Effective out $0.4195Token share0.9%
- Effective in $0.00838 / Effective out $1.1409Token share3.7%
- Effective in $0.00847 / Effective out $0.3361Token share11.6%
- Effective in $0.00891 / Effective out $0.6626Token share1.6%
- Effective in $0.00952 / Effective out $0.8398Token share4.1%
- Effective in $0.00963 / Effective out $0.4998Token share4.6%
- Effective in $0.01112 / Effective out $0.4417PriorityToken share1.7%
- Effective in $0.01115 / Effective out $0.5996Token share0.4%
- Effective in $0.01124 / Effective out $0.9707Token share22.2%
- Effective in $0.01165 / Effective out $0.4998Token share2.6%
- Effective in $0.01229 / Effective out $0.4997Token share6.8%
- Effective in $0.01244 / Effective out $0.7071Token share3.3%
- Effective in $0.01247 / Effective out $0.4998Token share2.7%
- Effective in $0.01470 / Effective out $0.6042Token share5.5%
- Effective in $0.01497 / Effective out $0.6044Token share0.7%
- Effective in $0.01546 / Effective out $0.7795Token share0.0%
- Effective in $0.01641 / Effective out $0.7195Token share3.4%
- Effective in $0.01688 / Effective out $0.4428Token share0.1%
- Effective in $0.01747 / Effective out $1.0996Token share0.1%
- Effective in $0.02003 / Effective out $0.6796Token share0.1%
- Effective in $0.02082 / Effective out $1.1996Token share6.7%
- Effective in $0.02346 / Effective out $1.0000Token share0.7%
- Effective in $0.02360 / Effective out $1.1996Token share0.9%
- Effective in $0.02645 / Effective out $0.7196Token share0.2%
- Effective in $0.02852 / Effective out $1.1996Token share1.2%
- Effective in $0.03141 / Effective out $1.1971Token share0.3%
- Effective in $0.03155 / Effective out $1.1994Token share0.0%
- Effective in $0.03171 / Effective out $0.8397Token share0.3%
- Effective in $0.03206 / Effective out $0.6196Token share0.8%
- Effective in $0.03207 / Effective out $1.1996Token share0.3%
- Effective in $0.03302 / Effective out $1.1996Token share1.3%
- Effective in $0.03547 / Effective out $0.4996Token share0.2%
- Effective in $0.03787 / Effective out $1.1997Token share0.3%
- Effective in $0.04296 / Effective out $1.1996Token share0.3%
- Effective in $0.04848 / Effective out $1.7992Token share0.3%
- Effective in $0.04975 / Effective out $0.9586Token share0.0%
- Effective in $0.06864 / Effective out $1.1291Token share0.0%
- Effective in $0.07900 / Effective out $1.1995Token share0.0%
- Effective in $0.08317 / Effective out $0.5495Token share0.0%
- Effective in $0.10301 / Effective out $2.3995FastToken share0.3%
- Effective in $0.18910 / Effective out $1.1997Token share0.1%
- Effective in $0.01086 / Effective out $0.1003Token share1.1%
- Effective in $0.01088 / Effective out $0.0996Token share0.8%
- Not used in Standard routing:Why these endpoints are not used
- Effective in $0.00453 / Effective out $0.2365FlexToken share1.0%
- Effective in $0.01340 / Effective out $0.2494FlexToken share2.3%
Performance
Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).
P50, best across providers
P50, best provider
Uptime
Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.
Availability over the last 3 days
Last 72 hoursAvailability over the last 24 hours
When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.
Benchmarks
Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.
| Source | Benchmark | Score |
|---|---|---|
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) Intelligence Index | 28.6 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) Terminal-Bench 4.0 | 28.8% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) AA-Omniscience Accuracy | 38.9% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (High) AA-Omniscience Non-Hallucination Rate | 38.8% |
| Artificial Analysis | DeepSeek V4.1 Flash (Max) HLE | 39.2% |
| Artificial Analysis | DeepSeek V4.1 Flash (Max) AA-LCR | 84.0% |
| Artificial Analysis | DeepSeek V4.1 Flash (Max) GDPval-AA | 55.0% |
| Artificial Analysis | DeepSeek V4.1 Flash (Max) CritPt | 14.3% |
| Artificial Analysis | DeepSeek V4.1 Flash (Max) SciCode | 51.9% |
| Artificial Analysis | DeepSeek V4.1 Flash (Max) Terminal-Bench 4.0 | 26.8% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) Intelligence Index | 18.5 |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) HLE | 11.5% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) AA-LCR | 57.3% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) CritPt | 0.3% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) SciCode | 38.5% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) Terminal-Bench 4.0 | 9.1% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) AA-Omniscience Accuracy | 21.9% |
| Artificial Analysis | DeepSeek V4.1 Flash 1017 (Non-reasoning) AA-Omniscience Non-Hallucination Rate | 74.7% |
| Design Arena | DeepSeek-V4.1-Flash (1017) Agents Arena Spreadsheet Elo | 1263 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena 3D Elo | 1343 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena Asciiart Elo | 1223 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena Code Categories Elo | 1287 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena Data Visualization Elo | 1273 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena Game Development Elo | 1251 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena SVG Elo | 1241 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena UI Component Elo | 1283 |
| Design Arena | DeepSeek-V4.1-Flash (1017) Models Arena Website Elo | 1295 |
| Vals AI | DeepSeek V4.1 Flash 1017 Access to Justice Benchmark | 73.1% |
| Vals AI | DeepSeek V4.1 Flash 1017 Code Migration v1.1 | 67.5% |
| Vals AI | DeepSeek V4.1 Flash 1017 CyberOffense | 54.8% |
| Vals AI | DeepSeek V4.1 Flash 1017 EMB v1.1 | 69.2% |
| Vals AI | DeepSeek V4.1 Flash 1017 Finance Agent v2.1 | 62.9% |
| Vals AI | DeepSeek V4.1 Flash 1017 Vals Finance Index | 60.9% |
| Vals AI | DeepSeek V4.1 Flash 1017 HLAB-Vals-1.1 | 14.2% |
| Vals AI | DeepSeek V4.1 Flash 1017 Vals Legal Index | 71.9% |
| Vals AI | DeepSeek V4.1 Flash 1017 Legal Research Bench | 45.2% |
| Vals AI | DeepSeek V4.1 Flash 1017 Tax Agent Bench | 44.4% |
| Vals AI | DeepSeek V4.1 Flash 1017 Terminal-Bench 4.0 | 31.8% |
| Vals AI | DeepSeek V4.1 Flash 1017 Vals Index | 61.3% |
| Vals AI | DeepSeek V4.1 Flash 1017 Vibe Code Bench 1-100 v1.1 | 26.2% |
| Vals AI | DeepSeek V4.1 Flash 1017 Vibe Code Bench v1.1 | 87.7% |
Apps
Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.
Activity
Token volume and request traffic to this model over time.
API
Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.
DeepSeek: DeepSeek Flash versions
Not currently available through EU in-region routing. In-region routing docs
Documentation
Guides, cookbooks, blog posts, and tools for building with DeepSeek: DeepSeek V4.1 Flash 1017 on OpenRouter.
Frequently asked questions
DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the deepseek-flash API model name. V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture.
The DeepSeek V4.1 Flash 1017 documentation on OpenRouter includes the Best OCR Model for Scanned Documents, Measured blog post, the Build an AI Code Reviewer with OpenRouter blog post, the Open-Weight vs Closed-Weight Models: What's the Difference? blog post, the How to Extract Tables from a PDF with an LLM blog post, and the Does DeepSeek V4 Have Vision? blog post. The quickstart covers API keys and your first request, and the API reference lists every parameter.
DeepSeek V4.1 Flash 1017 costs $0.036/M input tokens and $0.50/M output tokens, with separate rates for Cache Read at $0.01/M tokens.
DeepSeek V4.1 Flash 1017 has a 1,048,576 token context window. It supports up to 1,048,576 completion tokens.
Yes. DeepSeek V4.1 Flash 1017 accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.
DeepSeek V4.1 Flash 1017 accepts text and images as input and returns text.
DeepSeek V4.1 Flash 1017 is served by 38 providers on OpenRouter: Particle AI, Wafer, Prime Intellect, Sail Research, Open Inference, Morph, Ionstream, DeepInfra and 30 more. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.
DeepSeek V4.1 Flash 1017 was released on October 17, 2026.