Route order log in sign up
1 pick your primary route

DeepSeek: DeepSeek V4.1 Flash 1017

deepseek/deepseek-v4.1-flash-1017 · agent · text+image->text · 1M ctx · added 2026-10-17
$0.0145·0.6787 /M effective hit 93% 8.69T/d

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...

view
openrouter/deepseek/deepseek-v4.1-flash-1017
DeepSeek: DeepSeek V4.1 Flash 1017 · agent · text+image->text · 1M ctx
8.69T/d 93% $0.0145·0.6787 – 25d
sticker
life 429 tok/d dp age
¢ᵉ
deepinfra/fp8/flex
$0.0784·0.2352·0.0024 $0.0043·0.2352ⁱ $0.0043·0.2356ᵉ 97% – – 2.0% ✓ 3541ms 50t/s 1d ·
¢ⁱ
sail-research/fp4/us
$0.1010·0.1500·0.0067 $0.0092·0.1500ⁱ $0.0143·0.1494ᵉ 92% – – 0.1% ✓ 1184ms 76t/s 1d ·
particle-ai/us/flex
$0.0250·0.2500·0.0030 $0.0036·0.2500ⁱ $0.0075·0.2494ᵉ 77% – – 0.7% ✓ 667ms 100t/s 1d ·
ionstream/flex
$0.0825·0.2475·0.0052 $0.0073·0.2475ⁱ $0.0129·0.2468ᵉ 89% – – 0.3% ✓ 646ms 210t/s 1d ·
★
deepinfra/fp8
$0.0980·0.2940·0.0029 $0.0054·0.2940ⁱ $0.0077·0.3137ᵉ 95% – – 15% ✓ 745ms 96t/s 1d ·
morph/fp8
$0.0220·0.4000·0.0040 $0.0045·0.4000ⁱ $0.0062·0.3991ᵉ 88% – – 3.8% ✓ 1249ms 65t/s 1d ·
deepinfra/fp8/priority
$0.1470·0.4410·0.0044 $0.0081·0.4410ⁱ $0.0104·0.4426ᵉ 96% – – 1.9% ✓ 855ms 104t/s 1d ·
¢ ✓
relace/fp4
$0.0060·0.5000·0.0050 $0.0050·0.5000ⁱ $0.0051·0.4997ᵉ 93% – – 6.5% ✓ 693ms 90t/s 1d ·
streamlake/fp8
$0.1110·0.4440·0.0022 $0.0050·0.4440ⁱ $0.0114·0.4438ᵉ 91% – – 0.3% ◐ 2251ms 96t/s 1d ·
sail-research/fp4
$0.1515·0.2250·0.0101 $0.0138·0.2250ⁱ $0.0410·0.2242ᵉ 60% – – <0.1% ✓ 1124ms 67t/s 1d ·
river
$0.1200·0.4800·0.0024 $0.0054·0.4800ⁱ $0.0136·0.4795ᵉ 92% – – 0.2% ✓ 996ms 113t/s ✕11h ·
≈
wafer/mxfp4
$0.0660·0.5000·0.0065 $0.0080·0.5000ⁱ $0.0121·0.4997ᵉ 91% – – 2.9% ✓ 384ms 169t/s 1d ·
wafer/us
$0.0380·0.5000·0.0100 $0.0107·0.5000ⁱ $0.0123·0.4997ᵉ 91% – – 5.5% ✓ 389ms 171t/s 1d ·
decart/fp8
$0.1230·0.4920·0.0025 $0.0056·0.4920ⁱ $0.0143·0.4914ᵉ 90% – – 0.5% ✓ 1371ms 150t/s ✕2h ·
particle-ai/us
$0.0500·0.5000·0.0060 $0.0071·0.5000ⁱ $0.0156·0.4994ᵉ 71% – – 0.6% ✓ 629ms 103t/s 1d ·
stepfun/fp8
$0.1350·0.5400·0.0027 $0.0061·0.5400ⁱ $0.0131·0.5395ᵉ 93% – – 1.5% ◐ 749ms 123t/s 1d ·
cloudflare
$0.1000·0.6000·0.0030 $0.0055·0.6000ⁱ $0.0106·0.5995ᵉ 92% – – 1.0% ✓ 2420ms 77t/s 1d ·
ionstream
$0.1650·0.4950·0.0105 $0.0145·0.4950ⁱ $0.0259·0.4944ᵉ 91% – – 1.0% ✓ 948ms 136t/s 1d ·
inference-net/fp8
$0.0800·0.6050·0.0080 $0.0099·0.6050ⁱ $0.0153·0.5971ᵉ 91% – – 1.9% ✓ 235ms 111t/s 1d ·
inference-net/us
$0.0800·0.6050·0.0080 $0.0099·0.6050ⁱ $0.0154·0.6003ᵉ 91% – – 5.5% ✓ 254ms 121t/s 1d ·
primeintellect
$0.0437·0.7000·0.0024 $0.0035·0.7000ⁱ $0.0054·0.6995ᵉ 94% – – 1.6% ✓ 845ms 129t/s 1d ·
novita/fp8
$0.1650·0.6600·0.0033 $0.0075·0.6600ⁱ $0.0110·0.6597ᵉ 95% – – 2.5% ✓ 1437ms 81t/s 1d ·
siliconflow/fp8
$0.3000·0.5500·0.0060 $0.0136·0.5500ⁱ $0.0240·0.5494ᵉ 94% – – 0.1% ✓ 1211ms 64t/s 1d ·
cohere/fp8
$0.1500·0.5000·0.0030 $0.0068·0.5000ⁱ $0.0365·0.4997ᵉ 77% – – 1.3% ✓ 376ms 138t/s 1d ·
modelrun/fp4
$0.1700·0.6700·0.0060 $0.0102·0.6700ⁱ $0.0191·0.6695ᵉ 92% – – 0.3% ✓ 855ms 139t/s ✕12h ·
♻
atlas-cloud/fp8
$0.1950·0.7800·0.0039 $0.0088·0.7800ⁱ $0.0091·0.7797ᵉ 97% – – 0.1% ◐ 3544ms 77t/s 1d ·
digitalocean
$0.1800·0.7200·0.0060 $0.0105·0.7200ⁱ $0.0158·0.7194ᵉ 95% – – 4.5% ✓ 702ms 167t/s 1d ·
open-inference/fp4
$0.0700·0.8000·0.0060 $0.0077·0.8000ⁱ $0.0100·0.7998ᵉ 93% – – 1.6% ✓ 921ms 110t/s 1d ·
dekallm
$0.1800·0.6800·0.0060 $0.0105·0.6800ⁱ $0.0240·0.6795ᵉ 90% – – 0.3% ✓ 1217ms 86t/s 1d ·
≈ ◔
alibaba
$0.2100·0.8400·0.0042 $0.0095·0.8400ⁱ $0.0271·0.8394ᵉ 95% – – 1.1% ◐ 835ms 131t/s 1d ·
gmicloud/fp8
$0.1800·0.7200·0.0036 $0.0082·0.7200ⁱ $0.0445·0.7197ᵉ 80% – – 0.1% ◐ 979ms 133t/s 1d ·
≈ ◔
deepseek
$0.1500·0.6000·0.0030 $0.0068·0.6000ⁱ $0.0150·0.9904ᵉ 96% – – 13% ⚠ 1293ms 112t/s 1d ·
fireworks/spot
$0.2100·0.8400·0.0042 $0.0095·0.8400ⁱ $0.0411·0.8396ᵉ 82% – – 0.2% ✓ 1515ms 132t/s 1d ·
modal
$0.2500·1.0000·0.0050 $0.0113·1.0000ⁱ $0.0239·1.0000ᵉ 93% – – 1.0% ✓ 743ms 123t/s 1d ·
reka
$0.1600·1.1000·0.0080 $0.0119·1.1000ⁱ $0.0260·1.0996ᵉ 89% – – <0.1% ✓ 1031ms 95t/s 1d ·
phala
$0.2000·1.2000·0.0060 $0.0110·1.2000ⁱ $0.0173·1.1991ᵉ 74% – – <0.1% ✓ 1390ms 96t/s ✕12h ·
together
$0.3000·1.2000·0.0060 $0.0136·1.2000ⁱ $0.0218·1.1996ᵉ 95% – – 11% ✓ 517ms 200t/s 1d ·
primalabs
$0.2400·0.9600·0.0048 $0.0109·0.9600ⁱ $0.0498·0.9588ᵉ 85% – – <0.1% ✓ 1160ms 105t/s 1d ·
venice/fp8
$0.3000·1.2000·0.0075 $0.0151·1.2000ⁱ $0.0257·1.1995ᵉ 93% – – <0.1% ✓ 883ms 143t/s 1d ·
mistral/fp8
$0.3000·1.2000·0.0070 $0.0146·1.2000ⁱ $0.0263·1.1996ᵉ 93% – – 0.5% ✓ 1390ms 186t/s 1d ·
fireworks
$0.3000·1.2000·0.0060 $0.0136·1.2000ⁱ $0.0319·1.1997ᵉ 90% – – 2.2% ✓ 1012ms 170t/s 1d ·
crusoe/fp8
$0.2900·1.2000·0.0070 $0.0143·1.2000ⁱ $0.0320·1.1996ᵉ 90% – – 1.1% ✓ 584ms 189t/s 1d ·
makora/fp8
$0.1500·1.2000·0.0060 $0.0097·1.2000ⁱ $0.0324·1.1997ᵉ 81% – – 0.2% ✓ 1006ms 123t/s 1d ·
parasail/fp8
$0.3000·1.2000·0.0060 $0.0136·1.2000ⁱ $0.0325·1.1995ᵉ 93% – – 2.3% ✓ 756ms 185t/s 1d ·
baseten
$0.3000·1.2000·0.0070 $0.0146·1.2000ⁱ $0.0357·1.1996ᵉ 90% – – 0.2% ✓ 809ms 135t/s 1d ·
coreweave/fp8
$0.3000·1.2000·0.0060 $0.0136·1.2000ⁱ $0.0373·1.1995ᵉ 91% – – 1.4% ✓ 480ms 206t/s 1d ·
mistral
$0.3000·1.2000·0.0070 $0.0146·1.2000ⁱ $0.0477·1.1997ᵉ 88% – – <0.1% ✓ 1102ms 172t/s ✕10h ·
nebius/fp8
$0.3000·1.2000·0.0060 $0.0136·1.2000ⁱ $0.0643·1.1916ᵉ 79% – – <0.1% ✓ 2391ms 66t/s 1d ·
≈
baidu/fp8
$0.2800·1.1300·0.0060 $0.0131·1.1300ⁱ $0.0962·1.1030ᵉ 83% – – 0.2% ◐ 1225ms 76t/s 1d ·
fireworks/us
$0.4500·1.8000·0.0090 $0.0204·1.8000ⁱ $0.0507·1.7993ᵉ 93% – – 1.4% ✓ 524ms 171t/s 1d ·
near-ai
$0.3000·1.2000·0.0060 $0.0136·1.2000ⁱ $0.1273·1.1993ᵉ 65% – – <0.1% ✓ 845ms 32t/s 1d ·
⚡
baseten/fast
$0.6000·2.4000·0.0140 $0.0292·2.4000ⁱ $0.2636·2.3994ᵉ 57% – – <0.1% ✓ 286ms 534t/s 1d ·
vercel/deepseek/deepseek-v4.1-flash-1017
DeepSeek V4.1 Flash 1017 · agent · text+image->text · 1M ctx
– 93% $0.0215·0.9800 12% 25d
requesty/deepseek/deepseek-v4.1-flash-1017
deepseek-v4.1-flash-1017 · agent · text+image->text · 1M ctx
– 89% $0.0163·1.2000 21% 25d
hf/deepseek-ai/DeepSeek-V4.1-Flash-1017
DeepSeek-V4.1-Flash-1017 · agent · text+image->text · 1M ctx
– $0.3000·1.2000 – 25d