Gemini 3.5 Flash vs GPT-3.5 Turbo 16k
Prices checked
Gemini 3.5 Flash takes 3 of 5 rounds against GPT-3.5 Turbo 16k. It is also the cheaper of the two to run, at $1.69 per million tokens against $3.25.
GoogleWins 3 roundsGemini 3.5 Flash
1M context · tool calling · reads images
$1.69/M tokensCheapest price, at Kilo Gateway · $0.72 per 1,000 pages
OpenAIGPT-3.5 Turbo 16k
16k context · tool calling
$3.25/M tokensOpenAI's own price · $2.04 per 1,000 pages
Maker's price in (lower wins)
$1.50/M
$3/M
Maker's price out (lower wins)
$9/M
$4/M
Cheapest price (lower wins)
$1.69/M at Kilo Gateway
none below the maker
Maker's price, blended (lower wins)
$3.38/M
$3.25/M
Cost per 1,000 pages (lower wins)
$0.72
$2.04
Context
1M tokens
16k tokens
Open weights
no
not published
Tool calling
yes
yes
Structured output
yes
yes
Reads images
yes
no
Reasoning
yes
no
Released
19 May 2026
28 Aug 2023
Providers selling it
42
6
Price we measure against
the maker's own
the maker's own
Trains on your prompts (no is better)
not published
no provider to check
Providers selling both · 4
Kilo Gateway$1.69 / $3.25
FastRouter$3.38 / $3.25
Nous Portal$3.38 / $3.25
OpenRouter$3.38 / $3.25
Gemini 3.5 Flash, then GPT-3.5 Turbo 16k, per million tokens.
Go further
Add a third or fourth model, or change the weighting, in the tool.
Frequently asked
- Which is cheaper, Gemini 3.5 Flash or GPT-3.5 Turbo 16k?
- Gemini 3.5 Flash, at $1.69 per million tokens against $3.25 for GPT-3.5 Turbo 16k, blended 3 in : 1 out, read on 31 Oct 2026.
- Where is each one cheapest?
- Gemini 3.5 Flash: $1.69 at Kilo Gateway. GPT-3.5 Turbo 16k: nobody sells it below the maker's own price with a stated reason, so $3.25 is the number to beat.
- Can I buy both from one provider?
- Yes. 4 providers sell both: Kilo Gateway, FastRouter, Nous Portal, OpenRouter. The cheapest of them for the two together is Kilo Gateway.
- Which holds more context?
- Gemini 3.5 Flash, at 1M tokens against 16k.
Prices exclude tax, read 31 Oct 2026 from each provider's own catalog. They are per M tokens, blended 3 in : 1 out, and a page is 600 tokens in and 60 out. Bars are scaled to the larger value in each row.
Related comparisons