DeepSeek V4.1 Flash pricing & benchmarks
DeepSeek V4.1 Flash is available from DeepSeek at $0.300 per million input tokens and $1.20 per million output tokens ($0.525 blended at 3:1). It accepts up to 1,048,576 tokens of context. DeepSeek 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...
Pricing
Per million tokens · OpenRouter
Input$0.300
Output$1.20
Blended (3:1)$0.525
Specification
deepseek/deepseek-v4.1-flash
Context window1.0M
Max output944K
ReleasedSep 2026
LicenceOpen weights
Modalitiesin:text, in:image, out:text
Usage on OpenRouter
How much DeepSeek V4.1 Flash is actually being run — not how it benchmarks
Board rank
#1 of 15
Tokens, trailing week
42.2T
Change vs 7 days earlier
+12.7%
2026-09-222026-10-22
Each point is the trailing seven-day token total as of that day, which is what
OpenRouter's model leaderboard publishes — not that day's usage on its own. Sampled hourly by
token.app and stored since 2026-09-22; only the models currently on OpenRouter's board
appear here, so a missing panel means we have no current data, not zero usage.
No independently-run benchmark scores are published for DeepSeek V4.1 Flash yet. We show nothing rather than reprinting an unverified figure.
Where to run DeepSeek V4.1 Flash
38 options across 30 providers · 9.3× between cheapest and dearest. Blended at 3:1. ⚠ marks an option serving less context than the best available, or re-pricing above a prompt-length threshold. Bracketed labels are the provider's own service tier or region.
| Host | Blended $/1M | In / Out | Context | Weights | Uptime 24h |
|---|---|---|---|---|---|
| Decart | $0.113 | $0.090 / $0.180 | 1.0M | fp4 | 100.0% |
| Sail Research (fp4/us) | $0.130 | $0.090 / $0.250 | 1.0M | fp4 | 100.0% |
| Sail Research | $0.130 | $0.090 / $0.250 | 1.0M | fp4 | 100.0% |
| DeepInfra (fp8/flex) | $0.168 | $0.112 / $0.336 | 1.0M | fp8 | 89.7% |
| OpenInference | $0.177 | $0.070 / $0.500 | 1.0M | fp4 | 99.8% |
| Relace | $0.177 | $0.070 / $0.500 | 1.0M | — | 99.8% |
| DeepInfra | $0.210 | $0.140 / $0.420 | 1.0M | fp8 | 100.0% |
| AtlasCloud | $0.236 | $0.135 / $0.540 | 1.0M | fp8 | 100.0% |
| InferenceNet (us) | $0.256 | $0.103 / $0.715 | 1.0M | fp8 | 100.0% |
| InferenceNet | $0.256 | $0.103 / $0.715 | 1.0M | fp8 | 100.0% |
| StreamLake | $0.257 | $0.147 / $0.588 | 1.0M | fp8 | 100.0% |
| CoreWeave | $0.275 | $0.150 / $0.650 | 1.0M | fp8 | 100.0% |
| Reka | $0.289 | $0.105 / $0.840 | 1.0M | — | 94.6% |
| Wafer | $0.305 | $0.190 / $0.650 | 1.0M | mxfp4 | 100.0% |
| GMICloud | $0.315 | $0.180 / $0.720 | 1.0M | fp8 | 99.8% |
| DeepInfra (fp8/priority) | $0.315 | $0.210 / $0.630 | 1.0M | fp8 | 94.3% |
| Novita | $0.341 | $0.195 / $0.780 | 1.0M | fp8 | 99.9% |
| Fireworks (spot) | $0.367 | $0.210 / $0.840 | 1.0M | — | 100.0% |
| Phala | $0.367 | $0.210 / $0.840 | 1.0M | — | 100.0% |
| Morph | $0.385 | $0.313 / $0.600 | 1.0M | fp8 | 98.8% |
| DigitalOcean | $0.394 | $0.225 / $0.900 | 1.0M | — | 100.0% |
| Makora | $0.487 | $0.260 / $1.17 | 1.0M | fp8 | 99.9% |
| Crusoe | $0.517 | $0.290 / $1.20 | 1.0M | fp8 | 100.0% |
| Google (global) | $0.525 | $0.300 / $1.20 | 1.0M | — | 98.3% |
| Nebius | $0.525 | $0.300 / $1.20 | 1.0M | fp8 | 100.0% |
| PrimeIntellect | $0.525 | $0.300 / $1.20 | 1.0M | — | 100.0% |
| Baidu | $0.525 | $0.300 / $1.20 | 1.0M | fp8 | 100.0% |
| BaseTen (us) | $0.525 | $0.300 / $1.20 | 1.0M | — | 100.0% |
| Together | $0.525 | $0.300 / $1.20 | 1.0M | — | 100.0% |
| SiliconFlow | $0.525 | $0.300 / $1.20 | 1.0M | fp8 | 100.0% |
| Modal | $0.525 | $0.300 / $1.20 | 1.0M | — | 100.0% |
| BaseTen | $0.525 | $0.300 / $1.20 | 1.0M | — | 100.0% |
| Parasail | $0.525 | $0.300 / $1.20 | 1.0M | fp8 | 100.0% |
| Fireworks | $0.525 | $0.300 / $1.20 | 1.0M | — | 99.7% |
| Venice | $0.525 | $0.300 / $1.20 | 1M | fp8 | 100.0% |
| Alibaba | $0.578 | $0.330 / $1.32 | 1M | — | 100.0% |
| Fireworks (us) | $0.787 | $0.450 / $1.80 | 1.0M | — | 100.0% |
| BaseTen (fast) | $1.05 | $0.600 / $2.40 | 1.0M | — | 99.6% |
Host prices and uptime from OpenRouter, synced 2026-10-22. Lower-precision weights (fp4, fp8) can cost less and score worse than the same model at bf16 — the score above is not host-specific.
Compare DeepSeek V4.1 Flash
FAQ
How much does DeepSeek V4.1 Flash cost?
DeepSeek V4.1 Flash costs $0.300 per million input tokens and $1.20 per million output tokens on OpenRouter. At a 3:1 input:output mix that blends to $0.525 per million tokens.
What is DeepSeek V4.1 Flash's context window?
DeepSeek V4.1 Flash accepts up to 1,048,576 tokens of context, and can generate up to 943,718 output tokens.