DeepSeek V4.1 Flash 1017
- Input / output
- $0.03 / $0.25
- Providers
- 41
- Spread
- 6.8×
- 7 days
- −4.2%
- 30 days
- −24.4%
- Context
- 1M
Providers
| Provider | Input | Output | Blended | vs ref | Context | Uptime 1d | Price since |
|---|---|---|---|---|---|---|---|
| $0.03 | $0.25 | $0.085 | −45% | ||||
| fp8 | $0.05 | $0.30 | $0.113 | −27% | |||
DeepInfrafp8/flex fp8·−51% discount | $0.0784 | $0.235 | $0.118 | −24% | |||
| fp8·−51% discount | $0.098 | $0.294 | $0.147 | −5.0% | |||
Waferus mxfp4 | $0.036 | $0.50 | $0.152 | −1.8% | |||
| $0.09 | $0.36 | $0.158 | +1.8% | ||||
| $0.06 | $0.50 | $0.17 | +9.9% | ||||
| fp8 | $0.059 | $0.52 | $0.174 | +13% | |||
| mxfp4 | $0.066 | $0.50 | $0.175 | +13% | |||
| fp8·−65% discount | $0.105 | $0.42 | $0.184 | +19% | |||
| fp4 | $0.08 | $0.52 | $0.19 | +23% | |||
| fp8·−63% discount | $0.111 | $0.444 | $0.194 | +26% | |||
| $0.0437 | $0.70 | $0.208 | +34% | ||||
| fp8·−60% discount | $0.12 | $0.48 | $0.21 | +36% | |||
| $0.12 | $0.48 | $0.21 | +36% | ||||
| fp4 | $0.12 | $0.50 | $0.215 | +39% | |||
DeepInfrafp8/priority fp8·−51% discount | $0.147 | $0.441 | $0.221 | +42% | |||
| $0.10 | $0.60 | $0.225 | +45% | ||||
| $0.10 | $0.60 | $0.225 | +45% | ||||
| fp4 | $0.10 | $0.60 | $0.225 | +45% | |||
| fp8 | $0.08 | $0.666 | $0.227 | +46% | |||
| fp8 | $0.08 | $0.666 | $0.227 | +46% | |||
| fp8 | $0.15 | $0.50 | $0.238 | +53% | |||
| $0.15 | $0.60 | $0.263 | +70% | ||||
| fp8 | $0.15 | $0.60 | $0.263 | +70% | |||
| fp8·−47% discount | $0.15 | $0.605 | $0.263 | +70% | |||
| fp4 | $0.06 | $0.90 | $0.27 | +74% | |||
| fp8·−45% discount | $0.165 | $0.66 | $0.289 | +87% | |||
| fp4 | $0.17 | $0.67 | $0.295 | +91% | |||
| $0.21 | $0.62 | $0.313 | +102% | ||||
| fp8 | $0.15 | $0.80 | $0.313 | +102% | |||
| $0.18 | $0.72 | $0.315 | +104% | ||||
| fp8·−40% discount | $0.18 | $0.72 | $0.315 | +104% | |||
| fp4·−40% discount | $0.18 | $0.72 | $0.315 | +104% | |||
| fp8 | $0.195 | $0.78 | $0.341 | +121% | |||
Fireworksspot | $0.21 | $0.84 | $0.368 | +137% | |||
| fp8·−20% discount | $0.24 | $0.96 | $0.42 | +171% | |||
| $0.24 | $0.96 | $0.42 | +171% | ||||
| $0.25 | $1.00 | $0.438 | +183% | ||||
| fp8 | $0.29 | $1.20 | $0.518 | +234% | |||
| $0.30 | $1.20 | $0.525 | +239% | ||||
| fp8 | $0.30 | $1.20 | $0.525 | +239% | |||
| $0.30 | $1.20 | $0.525 | +239% | ||||
| $0.30 | $1.20 | $0.525 | +239% | ||||
| fp8 | $0.30 | $1.20 | $0.525 | +239% | |||
| $0.30 | $1.20 | $0.525 | +239% | ||||
| fp8 | $0.30 | $1.20 | $0.525 | +239% | |||
| $0.45 | $1.80 | $0.788 | +409% | ||||
BaseTenfast | $0.60 | $2.40 | $1.05 | +579% | |||
Morphfast fp8 | $0.60 | $2.40 | $1.05 | +579% |
- Particle AICheapestus/flex$0.085$0.03 in · $0.25 out−45% vs ref1M ctx99.9% uptime
- $0.113$0.05 in · $0.30 out−27% vs ref1M ctx92.5% uptimefp8
- fp8/flex$0.118$0.0784 in · $0.235 out−24% vs ref1M ctx94.1% uptimefp8·−51% discount
- $0.147$0.098 in · $0.294 out−5.0% vs ref1M ctx98.4% uptimefp8·−51% discount
- us$0.152$0.036 in · $0.50 out−1.8% vs ref1M ctx100% uptimemxfp4
- $0.158$0.09 in · $0.36 out+1.8% vs ref1M ctx100% uptime
- $0.17$0.06 in · $0.50 out+9.9% vs ref1M ctx100% uptime
- $0.174$0.059 in · $0.52 out+13% vs ref1M ctx99.7% uptimefp8
- $0.175$0.066 in · $0.50 out+13% vs ref1M ctx100% uptimemxfp4
- $0.184$0.105 in · $0.42 out+19% vs ref1M ctx100% uptimefp8·−65% discount
- $0.19$0.08 in · $0.52 out+23% vs ref1M ctx81.6% uptimefp4
- $0.194$0.111 in · $0.444 out+26% vs ref1M ctx100% uptimefp8·−63% discount
- $0.208$0.0437 in · $0.70 out+34% vs ref1M ctx100% uptime
- $0.21$0.12 in · $0.48 out+36% vs ref1M ctx100% uptimefp8·−60% discount
- $0.21$0.12 in · $0.48 out+36% vs ref1M ctx100% uptime
- $0.215$0.12 in · $0.50 out+39% vs ref1M ctx100% uptimefp4
- fp8/priority$0.221$0.147 in · $0.441 out+42% vs ref1M ctx98.4% uptimefp8·−51% discount
- $0.225$0.10 in · $0.60 out+45% vs ref1M ctx100% uptime
- $0.225$0.10 in · $0.60 out+45% vs ref1M ctx100% uptime
- $0.225$0.10 in · $0.60 out+45% vs ref1M ctx100% uptimefp4
- $0.227$0.08 in · $0.666 out+46% vs ref1M ctx100% uptimefp8
- $0.227$0.08 in · $0.666 out+46% vs ref1M ctx100% uptimefp8
- $0.238$0.15 in · $0.50 out+53% vs ref1M ctx100% uptimefp8
- $0.263$0.15 in · $0.60 out+70% vs ref1M ctx100% uptime
- $0.263$0.15 in · $0.60 out+70% vs ref1M ctx100% uptimefp8
- $0.263$0.15 in · $0.605 out+70% vs ref1M ctx99.9% uptimefp8·−47% discount
- $0.27$0.06 in · $0.90 out+74% vs ref1M ctx100% uptimefp4
- $0.289$0.165 in · $0.66 out+87% vs ref1M ctx100% uptimefp8·−45% discount
- $0.295$0.17 in · $0.67 out+91% vs ref1M ctx100% uptimefp4
- $0.313$0.21 in · $0.62 out+102% vs ref1M ctx98.7% uptime
- $0.313$0.15 in · $0.80 out+102% vs ref1M ctx97.1% uptimefp8
- $0.315$0.18 in · $0.72 out+104% vs ref1M ctx100% uptime
- $0.315$0.18 in · $0.72 out+104% vs ref1M ctx100% uptimefp8·−40% discount
- $0.315$0.18 in · $0.72 out+104% vs ref1M ctx100% uptimefp4·−40% discount
- $0.341$0.195 in · $0.78 out+121% vs ref1M ctx100% uptimefp8
- spot$0.368$0.21 in · $0.84 out+137% vs ref1M ctx89.3% uptime
- $0.42$0.24 in · $0.96 out+171% vs ref1M ctx100% uptimefp8·−20% discount
- $0.42$0.24 in · $0.96 out+171% vs ref1M ctx99.8% uptime
- $0.438$0.25 in · $1.00 out+183% vs ref1M ctx100% uptime
- $0.518$0.29 in · $1.20 out+234% vs ref1M ctx98.2% uptimefp8
- $0.525$0.30 in · $1.20 out+239% vs ref1M ctx100% uptime
- $0.525$0.30 in · $1.20 out+239% vs ref1M ctx100% uptimefp8
- $0.525$0.30 in · $1.20 out+239% vs ref1M ctx100% uptime
- $0.525$0.30 in · $1.20 out+239% vs ref1M ctx99.4% uptime
- $0.525$0.30 in · $1.20 out+239% vs ref1M ctx100% uptimefp8
- $0.525$0.30 in · $1.20 out+239% vs ref1M ctx100% uptime
- $0.525$0.30 in · $1.20 out+239% vs ref1M ctx100% uptimefp8
- $0.788$0.45 in · $1.80 out+409% vs ref1M ctx100% uptime
- fast$1.05$0.60 in · $2.40 out+579% vs ref1M ctx100% uptime
- fast$1.05$0.60 in · $2.40 out+579% vs ref1M ctx100% uptimefp8
Blended = (3 × input + output) / 4. Ref $0.155 = median of cheapest 20%.
Price history
USD per 1M tokens · each step is a price change
Prices in USD / 1M tokens.
Long-context pricing
| Offered by | Prompt size | Input | Output |
|---|---|---|---|
| Alibaba | base | $0.09 | $0.36 |
| > 0 | $0.18 | $0.72 | |
| > 0 | $0.09 | $0.36 | |
| DeepSeek | base | $0.15 | $0.60 |
| > 0 | $0.15 | $0.60 | |
| > 0 | $0.15 | $0.60 | |
| > 0 | $0.30 | $1.20 | |
| > 0 | $0.15 | $0.60 | |
| > 0 | $0.30 | $1.20 | |
| > 0 | $0.15 | $0.60 |
Other pricing
| Offered by | Cache read |
|---|---|
| Particle AI (us/flex), Pearl Research +7 | $0.003 |
| DeepInfra (fp8/flex) | $0.00235 |
| DeepInfra | $0.00294 |
| Wafer (us) | $0.01 |
| Alibaba | $0.0018 |
| Particle AI (us), Inceptron +7 | $0.006 |
| Morph, Modal | $0.005 |
| Wafer | $0.0065 |
| StepFun | $0.0021 |
| StreamLake | $0.00222 |
| PrimeIntellect, Ant Digital +1 | $0.0024 |
| DeepInfra (fp8/priority) | $0.00441 |
| InferenceNet, InferenceNet (us) | $0.008 |
| Baidu | $0.00321 |
| OpenInference | $0.002 |
| Novita | $0.0033 |
| Ionstream | $0.0105 |
| DigitalOcean, GMICloud +1 | $0.0036 |
| AtlasCloud | $0.0039 |
| Fireworks (spot) | $0.0042 |
| Mistral | $0.0056 |
| PrimaLabs | $0.0048 |
| Crusoe, BaseTen | $0.007 |
| Venice | $0.0075 |
| Fireworks (us) | $0.009 |
| BaseTen (fast), Morph (fast) | $0.014 |
Changes
- PriceDiscount price,60%→45%−25.0%
- PriceCache read price,$0.0024→$0.0033+37.5%
- PriceOutput price,$0.48→$0.66+37.5%
- PriceInput price,$0.12→$0.165+37.5%
- PriceDiscount price,—→47%—
- PriceCache read price,$0.006→$0.00321−46.5%
- PriceOutput price,$1.13→$0.605−46.5%
- PriceInput price,$0.28→$0.15−46.5%
- PriceOutput price,$0.80→$0.60−25.0%
- PriceInput price,$0.18→$0.10−44.4%
- PriceCache read price,$0.0036→$0.0018−50.0%
- PriceOutput price,$0.72→$0.36−50.0%
- PriceInput price,$0.18→$0.09−50.0%
- PriceOutput price,$0.635→$0.666+4.9%
- PriceOutput price,$0.635→$0.666+4.9%
About
The cheapest provider for DeepSeek V4.1 Flash 1017 is Particle AI at $0.085 per 1M tokens (blended), charging $0.03 per 1M input and $0.25 per 1M output tokens. 41 providers serve it; the most expensive charges 6.8× the typical price. Its best price fell 24.4% over the last 30 days. Context window: 1M tokens. Prices tracked since Oct 17, 2026.
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...
- Released
- Oct 17, 2026
- First tracked
- Oct 17, 2026
- Canonical slug
- deepseek/deepseek-v4.1-flash-20261017
- Weights
- deepseek-ai/DeepSeek-V4.1-Flash-1017