LLM prices per 1M tokens

264 models · 161 price changes

Gemini 3.8 Flash vs DeepSeek V4.1 Flash price

Input, output, cached and batch prices, every route and a month of tokens, side by side.

Gemini 3.8 FlashDeepSeek V4.1 Flash

A month of tokens

  • Gemini 3.8 Flash$150.00
  • DeepSeek V4.1 Flash$10.50

100M input and 20M output tokens, list price. Use your own numbers.

Side by side

Gemini 3.8 FlashDeepSeek V4.1 Flash
Input per 1M$0.75$0.045
Output per 1M$3.75$0.3
Cached input$0.075$0.015
Batch in / out$0.375 / $1.88n/a
Blended (3:1)$1.5$0.109
Context1M1M
Routes739
ReleasedSep 2, 2026Sep 10, 2026
Leadersboard score14 (#14)2 (#35)

On the price map

$0.01$0.03$0.1$0.3$1$3$10$30$0.03$0.1$0.3$1$3$10$30$100Input $ per 1M tokensOutput $ per 1M tokensxxMOOOOOOOOOOOOOOOOOOOOZMDMNGMMZDGMMZDMZGDQMDGGMMZGGDDNGGGGGGGGOOOQAAZQAQQAQQAAAQAQxQQAQQAQAAAAQQMxxxOOOOOOOOOOOOOOOOOAAONMNMQQMGGMQQQQQQMNMMMMMQQQQQMQQQQxMMCZZZZZZZAPAAOOPCDDDDDMDDGGMGGNZMOMOOZZMMQGGGMMMMMMOOOQMPPQCAOCCOOOOOMPGGGGZMMQQMGMMQQQ
Size: context window, 8K to 2MEqual blended cost (3:1) 247 models, lower left is cheaper

Every route

Gemini 3.8 Flash

Way to buyIn / out per 1M
Google Vertex AI
global-flex, cheapest · cached $0.037
$0.375 / $1.88-50% vs list
Google Gemini API
flex · cached $0.037
$0.375 / $1.88-50% vs list
Google Gemini API API
The lab's own API · cached $0.075
$0.75 / $3.75List price
Google Vertex AI
global · cached $0.075
$0.75 / $3.75Same as list
Google Vertex AI
us · cached $0.083
$0.825 / $4.13+10% vs list
Google Vertex AI
global-priority · cached $0.135
$1.35 / $6.75+80% vs list
Google Gemini API
priority · cached $0.135
$1.35 / $6.75+80% vs list

DeepSeek V4.1 Flash

Way to buyIn / out per 1M
InferenceNet
fp8 · cached $0.015
$0.045 / $0.3Same as list
Relace
fp4 · cached $0.005
$0.035 / $0.5+39% vs list
Wafer
us, mxfp4 · cached $0.006
$0.045 / $0.5+46% vs list
OpenInference
fp4 · cached $0.01
$0.1 / $0.5+84% vs list
Morph
fp8 · cached $0.0023
$0.12 / $0.468+90% vs list
InferenceNet
us, fp8 · cached $0.008
$0.103 / $0.6+109% vs list
AtlasCloud
fp8 · cached $0.013
$0.135 / $0.54+117% vs list
Wafer
mxfp4 · cached $0.05
$0.1 / $0.65+118% vs list
DeepInfra
fp8-flex, fp8 · cached $0.0048
$0.16 / $0.48+121% vs list
Mancer 2
fp8
$0.15 / $0.75+176% vs list
DeepInfra
fp8 · cached $0.006
$0.2 / $0.6+176% vs list
CoreWeave
fp8 · cached $0.02
$0.2 / $0.65+187% vs list
Fireworks AI
spot · cached $0.0042
$0.21 / $0.84+238% vs list
DigitalOcean
Default region · cached $0.006
$0.225 / $0.9+262% vs list
Reka
Default region · cached $0.01
$0.15 / $1.2+279% vs list
PrimeIntellect
Default region · cached $0.0048
$0.156 / $1.2+283% vs list
DeepInfra
fp8-priority, fp8 · cached $0.009
$0.3 / $0.9+314% vs list
Relace
fast, fp4 · cached $0.008
$0.3 / $0.9+314% vs list
Makora
fp8 · cached $0.0058
$0.26 / $1.17+348% vs list
Crusoe
fp8 · cached $0.007
$0.29 / $1.2+376% vs list
Baidu
fp8 · cached $0.006
$0.3 / $1.2+383% vs list
Baseten
Default region · cached $0.007
$0.3 / $1.2+383% vs list
Baseten
us · cached $0.007
$0.3 / $1.2+383% vs list
Cohere
fp8 · cached $0.003
$0.3 / $1.2+383% vs list
Fireworks AI
Default region · cached $0.006
$0.3 / $1.2+383% vs list
GMICloud
fp8 · cached $0.006
$0.3 / $1.2+383% vs list
Google Vertex AI
global · cached $0.03
$0.3 / $1.2+383% vs list
Modal
Default region · cached $0.03
$0.3 / $1.2+383% vs list
Nebius
fp8 · cached $0.01
$0.3 / $1.2+383% vs list
Novita AI
fp8 · cached $0.006
$0.3 / $1.2+383% vs list
Parasail
fp8 · cached $0.006
$0.3 / $1.2+383% vs list
Phala
Default region · cached $0.006
$0.3 / $1.2+383% vs list
SiliconFlow
fp8 · cached $0.006
$0.3 / $1.2+383% vs list
StreamLake
fp8 · cached $0.006
$0.3 / $1.2+383% vs list
Together AI
Default region · cached $0.006
$0.3 / $1.2+383% vs list
Venice
fp8 · cached $0.0075
$0.3 / $1.2+383% vs list
Qwen (Alibaba)
Default region · cached $0.033
$0.33 / $1.32+431% vs list
Fireworks AI
us · cached $0.009
$0.45 / $1.8+624% vs list
Parasail
fast, fp8 · cached $0.009
$0.45 / $1.8+624% vs list