DeepSeek V4 Flash 0731 vs GPT-4o mini.
model vs model · prices as of
DeepSeek V4 Flash 0731 takes 2 of 5 rounds against GPT-4o mini. It is also the cheaper of the two to run, at $0.25 per million tokens against $0.95.
DeepSeekwins 2 roundsDeepSeek V4 Flash 0731
$0.25/M tokens
cheapest we can explain, at LLM Gateway · $0.036 per 1,000 pages
1M context · tool calling · open weights
OpenAIGPT-4o mini
$0.95/M tokens
cheapest we can explain, at Poe · $0.115 per 1,000 pages
128k context · tool calling · reads images
List price in (lower wins)
$0.15/M
$0.15/M
List price out (lower wins)
$0.60/M
$0.60/M
Cheapest we can explain (lower wins)
$0.25/M at LLM Gateway
$0.95/M at Poe
Standard price (lower wins)
$1.05/M
$1.05/M
Cost per 1,000 pages (lower wins)
$0.036
$0.115
Context
1M tokens
128k tokens
Open weights
yes
no
Tool calling
yes
yes
Structured output
yes
yes
Reads images
no
yes
Reasoning
yes
no
Released
31 Jul 2026
18 Jul 2024
Providers selling it
97
30
Price we measure against
the maker's own
the maker's own
Trains on your prompts (no is better)
not stated
not stated
Providers selling both · 25
LLM Gateway$0.25 / $1.05
Kilo Gateway$0.34 / $1.05
Nous Portal$0.34 / $1.05
OpenRouter$0.34 / $1.05
NanoGPT$0.35 / $1.05
Cortecs$0.35 / $1.15
FastRouter$0.45 / $1.05
TrustedRouter$0.39 / $1.11
DeepSeek V4 Flash 0731 then GPT-4o mini, per million tokens · showing the 8 cheapest of 25
Go further
Add a third or fourth model, or switch the weighting, in the tool.
What moved in the last seven days
- DeepSeek V4 Flash 0731
DeepInfra$0.09 / $0.18 $0.06 / $0.1820% cheaper · - DeepSeek V4 Flash 0731
Inceptron$0.06 / $0.20 $0.06 / $0.188% cheaper · - DeepSeek V4 Flash 0731Inceptron$0.06 / $0.20 $0.06 / $0.188% cheaper ·
- DeepSeek V4 Flash 0731
io.net Intelligence$0.21 / $0.59 $0.18 / $0.5511% cheaper · - DeepSeek V4 Flash 0731Kilo Gateway$0.03 / $0.13 $0.06 / $0.1856% dearer ·
- DeepSeek V4 Flash 0731Nous Portal$0.03 / $0.13 $0.06 / $0.1856% dearer ·
- DeepSeek V4 Flash 0731OpenRouter$0.03 / $0.13 $0.06 / $0.1856% dearer ·
- DeepSeek V4 Flash 0731
Requesty$0.08 / $0.15 $0.28 / $0.56267% dearer · - DeepSeek V4 Flash 0731
StreamLake$0.06 / $0.17 $0.06 / $0.184% dearer · - DeepSeek V4 Flash 0731
Charm Hyper$0.20 / $0.40 $0.44 / $1.32164% dearer · - DeepSeek V4 Flash 0731Cortecs$0.09 / $0.18 $0.06 / $0.1824% cheaper ·
- GPT-4o miniRequesty$0.17 / $0.66 $0.15 / $0.609% cheaper ·
- DeepSeek V4 Flash 0731above.dev$0.24 / $0.73 $0.17 / $0.6620% cheaper ·
- DeepSeek V4 Flash 0731
GMI Cloud$0.09 / $0.18 $0.29 / $0.86277% dearer · - DeepSeek V4 Flash 0731
Mancer$0.17 / $0.50 $0.20 / $0.6021% dearer · - DeepSeek V4 Flash 0731DeepSeekDeepSeek's list price, in and out per million tokens$0.20 / $0.40 $0.15 / $0.605% dearer ·
- DeepSeek V4 Flash 0731
Morph$0.12 / $0.35 $0.14 / $0.4015% dearer · - DeepSeek V4 Flash 0731
Relace$0.07 / $0.18 $0.06 / $0.1220% cheaper · - DeepSeek V4 Flash 0731
Sail Research$0.07 / $0.30 $0.07 / $0.3414% dearer · - GPT-4o miniNous Portal$0.12 / $0.48 $0.15 / $0.6025% dearer ·
- DeepSeek V4 Flash 0731DeepSeek$0.14 / $0.28 $0.15 / $0.6050% dearer ·
- DeepSeek V4 Flash 0731
Merge Gateway$0.22 / $0.66 $0.14 / $0.2847% cheaper · - DeepSeek V4 Flash 0731
Ofox$0.44 / $1.32 $0.19 / $0.5159% cheaper · - DeepSeek V4 Flash 0731
Zenifra$2.40 / $6.90 $2.20 / $6.606% cheaper ·
Frequently asked
- Which is cheaper, DeepSeek V4 Flash 0731 or GPT-4o mini?
- DeepSeek V4 Flash 0731, at $0.25 per million tokens against $0.95 for GPT-4o mini, three parts input to one part output, read on 18 Sept 2026.
- Where is each one cheapest?
- DeepSeek V4 Flash 0731: $0.25 at LLM Gateway. GPT-4o mini: $0.95 at Poe.
- Can I buy both from one provider?
- Yes. 25 providers sell both: LLM Gateway, Kilo Gateway, Nous Portal, OpenRouter, NanoGPT, Cortecs, and more. The cheapest of them for the two together is LLM Gateway.
- Which holds more context?
- DeepSeek V4 Flash 0731, at 1M tokens against 128k.
Prices exclude tax and are per million tokens, read 18 Sept 2026 from each provider's own catalog. The standard price blends three parts input to one part output; a page is 600 tokens in and 60 out. Bars are scaled to the larger value in each row.