Pricingby GlyphDeepSeek 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.
Ways to buy 39
| Way to buy | Input | Output | In / out per 1M | Cached | Vs cheapest |
|---|---|---|---|---|---|
| InferenceNet fp8 · cached $0.015 | $0.045 | $0.3 | $0.045 / $0.3Cheapest | $0.015 | Cheapest |
| Relace fp4 · cached $0.005 | $0.035 | $0.5 | $0.035 / $0.5+39% vs cheapest | $0.005 | +39% vs cheapest |
| Wafer us, mxfp4 · cached $0.006 | $0.045 | $0.5 | $0.045 / $0.5+46% vs cheapest | $0.006 | +46% vs cheapest |
| OpenInference fp4 · cached $0.01 | $0.1 | $0.5 | $0.1 / $0.5+84% vs cheapest | $0.01 | +84% vs cheapest |
| Morph fp8 · cached $0.0023 | $0.12 | $0.468 | $0.12 / $0.468+90% vs cheapest | $0.0023 | +90% vs cheapest |
| InferenceNet us, fp8 · cached $0.008 | $0.103 | $0.6 | $0.103 / $0.6+109% vs cheapest | $0.008 | +109% vs cheapest |
| AtlasCloud fp8 · cached $0.013 | $0.135 | $0.54 | $0.135 / $0.54+117% vs cheapest | $0.013 | +117% vs cheapest |
| Wafer mxfp4 · cached $0.05 | $0.1 | $0.65 | $0.1 / $0.65+118% vs cheapest | $0.05 | +118% vs cheapest |
fp8-flex, fp8 · cached $0.0048 | $0.16 | $0.48 | $0.16 / $0.48+121% vs cheapest | $0.0048 | +121% vs cheapest |
| Mancer 2 fp8 | $0.15 | $0.75 | $0.15 / $0.75+176% vs cheapest | n/a | +176% vs cheapest |
fp8 · cached $0.006 | $0.2 | $0.6 | $0.2 / $0.6+176% vs cheapest | $0.006 | +176% vs cheapest |
fp8 · cached $0.02 | $0.2 | $0.65 | $0.2 / $0.65+187% vs cheapest | $0.02 | +187% vs cheapest |
spot · cached $0.0042 | $0.21 | $0.84 | $0.21 / $0.84+238% vs cheapest | $0.0042 | +238% vs cheapest |
Default region · cached $0.006 | $0.225 | $0.9 | $0.225 / $0.9+262% vs cheapest | $0.006 | +262% vs cheapest |
| Reka Default region · cached $0.01 | $0.15 | $1.2 | $0.15 / $1.2+279% vs cheapest | $0.01 | +279% vs cheapest |
| PrimeIntellect Default region · cached $0.0048 | $0.156 | $1.2 | $0.156 / $1.2+283% vs cheapest | $0.0048 | +283% vs cheapest |
fp8-priority, fp8 · cached $0.009 | $0.3 | $0.9 | $0.3 / $0.9+314% vs cheapest | $0.009 | +314% vs cheapest |
| Relace fast, fp4 · cached $0.008 | $0.3 | $0.9 | $0.3 / $0.9+314% vs cheapest | $0.008 | +314% vs cheapest |
| Makora fp8 · cached $0.0058 | $0.26 | $1.17 | $0.26 / $1.17+348% vs cheapest | $0.0058 | +348% vs cheapest |
fp8 · cached $0.007 | $0.29 | $1.2 | $0.29 / $1.2+376% vs cheapest | $0.007 | +376% vs cheapest |
| Baidu fp8 · cached $0.006 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.006 | +383% vs cheapest |
Default region · cached $0.007 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.007 | +383% vs cheapest |
us · cached $0.007 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.007 | +383% vs cheapest |
fp8 · cached $0.003 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.003 | +383% vs cheapest |
Default region · cached $0.006 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.006 | +383% vs cheapest |
| GMICloud fp8 · cached $0.006 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.006 | +383% vs cheapest |
global · cached $0.03 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.03 | +383% vs cheapest |
Default region · cached $0.03 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.03 | +383% vs cheapest |
fp8 · cached $0.01 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.01 | +383% vs cheapest |
fp8 · cached $0.006 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.006 | +383% vs cheapest |
fp8 · cached $0.006 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.006 | +383% vs cheapest |
| Phala Default region · cached $0.006 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.006 | +383% vs cheapest |
fp8 · cached $0.006 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.006 | +383% vs cheapest |
| StreamLake fp8 · cached $0.006 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.006 | +383% vs cheapest |
Default region · cached $0.006 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.006 | +383% vs cheapest |
fp8 · cached $0.0075 | $0.3 | $1.2 | $0.3 / $1.2+383% vs cheapest | $0.0075 | +383% vs cheapest |
Default region · cached $0.033 | $0.33 | $1.32 | $0.33 / $1.32+431% vs cheapest | $0.033 | +431% vs cheapest |
us · cached $0.009 | $0.45 | $1.8 | $0.45 / $1.8+624% vs cheapest | $0.009 | +624% vs cheapest |
fast, fp8 · cached $0.009 | $0.45 | $1.8 | $0.45 / $1.8+624% vs cheapest | $0.009 | +624% vs cheapest |
USD per 1M tokens. Regions are separate routes.
Through a gateway
Your own numbers100M input and 20M output tokens a month: $10.50 direct.
Kong AI Gateway$10.50+0%
LiteLLM$10.50+0%
Neon AI Gateway$10.50+0%
Vercel AI Gateway$10.50+0%
Requesty$11.03+5%
Eden AI$11.08+5.5%
OpenRouter$11.08+5.5%
Cloudflare AI GatewayNot offeredNot offered
HeliconeNot offeredNot offered
TrueFoundry AI GatewayNot offeredNot offered
A dash: not on that gateway's own model list. No published token fee: Martian Gateway, Portkey.
Related models
| In / out per 1M | Cheapest route | |||||||
|---|---|---|---|---|---|---|---|---|
DeepSeek · 1MCheapest: $0.081 on Particle AI (us-flex) | $0.15 | $0.6 | $0.15 / $0.6 | $0.262 | 1M | $0.081 on Particle AI (us-flex) | · | 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. |
DeepSeek · 1M | $0.44 | $1.32 | $0.44 / $1.32 | $0.66 | 1M | on DeepInfra (fp8) | · | DeepSeek V4 Flash Vision Exp is an experimental vision-enabled version of DeepSeek V4 Flash 0731 from DeepSeek, adding image understanding while matching the base model on text capabi |
DeepSeek · 1M | $0.66 | $1.98 | $0.66 / $1.98 | $0.99 | 1M | Lab list price | · | DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. |
DeepSeek · 1M | $0.019 | $0.3 | $0.019 / $0.3 | $0.089 | 1M | on Sail Research (us) | · | DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. |
Anthropic · 1M | $0.1 | $0.5 | $0.1 / $0.5 | $0.2 | 1M | Lab list price | · | Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use. |
Cohere · 128K | $0.08 | Free | $0.08 / Free | $0.06 | 128K | Lab list price | · | Embed 5 Fast is the lower-latency, cost-efficient tier of Cohere's Embed 5 family of multimodal embedding models (about 1B parameters). |
Cohere · 128K | $0.12 | Free | $0.12 / Free | $0.09 | 128K | Lab list price | · | Embed 5 Pro is the higher-accuracy tier of Cohere's Embed 5 family of multimodal embedding models (about 2.4B parameters). |
 Titan Embeddings G1 - TextAmazon Nova · 8K | $0.1 | Free | $0.1 / Free | $0.075 | 8K | Lab list price | · | Amazon Titan Embeddings G1 - Text is the first-generation Titan text embedding model. |
About DeepSeek V4.1 Flash
Takes text, image in and returns text. First listed Sep 23, 2026. Up to 944K output tokens. Blended price $0.109 per 1M tokens at 3 input to 1 output. Overall score 2 on leadersboard.fru.dev (rank 35). Model id deepseek-v4-1-flash.
Price history
List price since Sep 23, 2026, from daily reads of the OpenRouter catalog.
DeepSeek V4.1 Flash
PrimeIntellect cut its price: input $0.3 to $0.156 per 1M tokens.
Mancer 2 (fp8) cut its price: input $0.2 to $0.15 and output $1 to $0.75 per 1M tokens.
DigitalOcean raised its price: input $0.165 to $0.225 and output $0.66 to $0.9 per 1M tokens.
InferenceNet (us) cut its price: output $0.715 to $0.6 per 1M tokens.
AtlasCloud (fp8) cut its price: input $0.3 to $0.135 and output $1.2 to $0.54 per 1M tokens.
PrimeIntellect cut its price: cached input $0.03 to $0.006 per 1M tokens.
Nebius (fp8) set its first price: cached input $0.01 per 1M tokens.
Sail Research (fp4) cut its price: output $0.3 to $0.25 per 1M tokens.
Sail Research (fp4-us) cut its price: input $0.101 to $0.09 and output $0.15 to $0.25 per 1M tokens.
Questions
How much does DeepSeek V4.1 Flash cost?
$0.045 per million input tokens and $0.3 per million output tokens, with cached input at $0.015.
What is the context window of DeepSeek V4.1 Flash?
1M tokens, with up to 944K tokens of output.
Where is DeepSeek V4.1 Flash cheapest?
Of 39 routes tracked, InferenceNet is cheapest at $0.045 input and $0.3 output per million tokens.
Has DeepSeek V4.1 Flash's price changed?
Yes. The input price went from $0.3 to $0.156, first seen Nov 3, 2026.