Gemini 2.5 Flash-Lite vs Llama Guard 4 12B
Prices checked
Gemini 2.5 Flash-Lite takes 4 of 5 rounds against Llama Guard 4 12B. It is also the cheaper of the two to run, at $0.09 per million tokens against $0.14.
GoogleWins 4 roundsGemini 2.5 Flash-Lite
1M context · tool calling · reads images
$0.09/M tokensCheapest price, at Kilo Gateway · 4.2¢ per 1,000 pages
MetaLlama Guard 4 12B
164k context · reads images · open weights
$0.14/M tokensCheapest price, at Kilo Gateway · 9.5¢ per 1,000 pages
Maker's price in (lower wins)
$0.10/M
$0.18/M
Maker's price out (lower wins)
$0.40/M
$0.18/M
Cheapest price (lower wins)
$0.09/M at Kilo Gateway
$0.14/M at Kilo Gateway
Maker's price, blended (lower wins)
$0.18/M
$0.18/M
Cost per 1,000 pages (lower wins)
4.2¢
9.5¢
Context
1M tokens
164k tokens
Open weights
no
yes
Tool calling
yes
no
Structured output
yes
no
Reads images
yes
yes
Reasoning
yes
no
Released
17 Jun 2025
30 Apr 2025
Providers selling it
28
6
Price we measure against
the maker's own
OpenRouter
Trains on your prompts (no is better)
not published
not published
Providers selling both · 4
Kilo Gateway$0.09 / $0.14
Eden AI$0.17 / $0.18
OpenRouter$0.17 / $0.18
TrustedRouter$0.18 / $0.19
Gemini 2.5 Flash-Lite, then Llama Guard 4 12B, per million tokens.
Go further
Add a third or fourth model, or change the weighting, in the tool.
Frequently asked
- Which is cheaper, Gemini 2.5 Flash-Lite or Llama Guard 4 12B?
- Gemini 2.5 Flash-Lite, at $0.09 per million tokens against $0.14 for Llama Guard 4 12B, blended 3 in : 1 out, read on 31 Oct 2026.
- Where is each one cheapest?
- Gemini 2.5 Flash-Lite: $0.09 at Kilo Gateway. Llama Guard 4 12B: $0.14 at Kilo Gateway.
- Can I buy both from one provider?
- Yes. 4 providers sell both: Kilo Gateway, Eden AI, OpenRouter, TrustedRouter. The cheapest of them for the two together is Kilo Gateway.
- Which holds more context?
- Gemini 2.5 Flash-Lite, at 1M tokens against 164k.
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