Which is cheaper, Gemma 4 31B IT or Llama-3.3-70B-Instruct?

Llama-3.3-70B-Instruct, at $0.38 per million tokens against $0.61 for Gemma 4 31B IT, three parts input to one part output, read on 16 Sept 2026.

Where is each one cheapest?

Gemma 4 31B IT: nobody goes below OpenRouter's own price with a reason we can point at, so $0.61 is the number to beat. Llama-3.3-70B-Instruct: $0.38 at NanoGPT.

Can I buy both from one provider?

Yes. 23 providers sell both: NanoGPT, DeepInfra, Kilo Gateway, Nous Portal, OpenRouter, LLM Gateway, and more. The cheapest of them for the two together is NanoGPT.

Which holds more context?

Gemma 4 31B IT, at 262k tokens against 128k.

Gemma 4 31B IT vs Llama-3.3-70B-Instruct.

model vs model · prices as of 16 Sept 2026

Gemma 4 31B IT takes 3 of 5 rounds against Llama-3.3-70B-Instruct. Llama-3.3-70B-Instruct counters on price: $0.38 per million tokens against $0.61, for 128k context instead of 262k.

Googlewins 3 rounds

Gemma 4 31B IT

$0.61/M tokens

OpenRouter list price, nothing below it is accounted for · $0.0744 per 1,000 pages

262k context · tool calling · reads images · open weights

Llama-3.3-70B-Instruct

$0.38/M tokens

cheapest we can explain, at NanoGPT · $0.0438 per 1,000 pages

128k context · tool calling · open weights

NanoGPT ↗

List price in (lower wins)

$0.09/M

$0.10/M

List price out (lower wins)

$0.34/M

$0.32/M

Cheapest we can explain (lower wins)

none below list

$0.38/M at NanoGPT

Standard price (lower wins)

$0.61/M

$0.62/M

Cost per 1,000 pages (lower wins)

$0.0744

$0.0438

Context

262k tokens

128k tokens

Open weights

yes

Tool calling

yes

Structured output

yes

not stated

Reads images

yes

Reasoning

yes

Released

2 Apr 2026

6 Dec 2024

Providers selling it

Price we measure against

OpenRouter

Trains on your prompts (no is better)

no provider to check

not stated

Providers selling both · 23

NanoGPT$0.35 / $0.38
DeepInfra$0.61 / $0.62
Kilo Gateway$0.61 / $0.62
Nous Portal$0.61 / $0.62
OpenRouter$0.61 / $0.62
LLM Gateway$0.55 / $0.81
Meganova$0.77 / $0.60
TrustedRouter$0.64 / $0.85

Gemma 4 31B IT then Llama-3.3-70B-Instruct, per million tokens · showing the 8 cheapest of 23

Go further

open in the compare tool →every price for Gemma 4 31B IT every price for Llama-3.3-70B-Instruct

Add a third or fourth model, or switch the weighting, in the tool.

What moved in the last seven days

Gemma 4 31B ITNous Portal$0.07 / $0.27 $0.09 / $0.3425% dearer · 16 Sept 2026
Llama-3.3-70B-InstructCortecs$0.13 / $0.41 $0.75 / $0.75268% dearer · 16 Sept 2026
Llama-3.3-70B-InstructEden AI$0.76 / $0.76 $0.75 / $0.750.7% cheaper · 16 Sept 2026
Llama-3.3-70B-InstructNous Portal$0.08 / $0.26 $0.10 / $0.3225% dearer · 16 Sept 2026
Llama-3.3-70B-InstructRequesty$0.13 / $0.40 $0.23 / $0.4038% dearer · 16 Sept 2026
Gemma 4 31B ITLLM Gateway$0.10 / $0.30 $0.10 / $0.259% cheaper · 11 Sept 2026
Gemma 4 31B ITRequesty$0.40 / $0.60 $0.13 / $0.3857% cheaper · 11 Sept 2026

Frequently asked

Which is cheaper, Gemma 4 31B IT or Llama-3.3-70B-Instruct?: Llama-3.3-70B-Instruct, at $0.38 per million tokens against $0.61 for Gemma 4 31B IT, three parts input to one part output, read on 16 Sept 2026.
Where is each one cheapest?: Gemma 4 31B IT: nobody goes below OpenRouter's own price with a reason we can point at, so $0.61 is the number to beat. Llama-3.3-70B-Instruct: $0.38 at NanoGPT.
Can I buy both from one provider?: Yes. 23 providers sell both: NanoGPT, DeepInfra, Kilo Gateway, Nous Portal, OpenRouter, LLM Gateway, and more. The cheapest of them for the two together is NanoGPT.
Which holds more context?: Gemma 4 31B IT, at 262k tokens against 128k.

Prices exclude tax and are per million tokens, read 16 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.