model comparison

vs

Gemma 4 31B vs GPT 5.6 Luna

Google AI against OpenAI, compared on listed price, real workload cost, context and capabilities.

the short answer

  • Gemma 4 31B lists the lower output price; input prices match.
  • GPT 5.6 Luna offers the larger context window.
  • GPT 5.6 Luna retired on 6 October 2026; its numbers are kept for reference.

side-by-side specs

specGemma 4 31BGPT 5.6 Luna
providerGoogle AIOpenAI
input price$0.2/M$0.2/M
output price$0.2/M$1.2/M
cached input—$0.02/M
blended price (3:1)$0.2/M$0.45/M
context window128k tokens1M tokens
max output—128k tokens
inputtext, imagestext, images
structured outputyesyes
weightsopen (Apache 2.0)closed
releasedApril 2, 2026—
availabilityliveretired 6 October 2026

same work, both models

workloadGemma 4 31BGPT 5.6 Luna
one chat reply3,000 tokens in, 500 outGPT 5.6 Luna costs 1.7× as much as Gemma 4 31B$0.0007$0.0012
summarise a 100-page report50,000 tokens in, 1,500 out, one requestGPT 5.6 Luna costs 1.1× as much as Gemma 4 31B$0.01$0.01
1,000 classification calls500 tokens in, 20 out, per callGPT 5.6 Luna costs 1.2× as much as Gemma 4 31B$0.10$0.12
coding-agent session2M tokens in across ~60 turns, 60,000 out, no cacheGPT 5.6 Luna costs 1.1× as much as Gemma 4 31B$0.41$0.47

at listed rates. token counts are typical sizes, not measurements.

Gemma 4 31B pricing notes

  • output bills at the same $0.20 per 1M tokens as input, so long generations cost no more per token than prompts.
  • no cached-input discount: repeated prompt tokens bill at the full input rate.

GPT 5.6 Luna pricing notes

  • requests over 272K input tokens billed at 2x input and 1.5x output for the whole request; the listed price is the standard rate.

when to pick which

Gemma 4 31B

Gemma 4 31B is the largest dense model in Google's Apache 2.0 open-weights Gemma 4 family, released April 2026. it ships alongside the 26B A4B mixture-of-experts model, takes text and images, and has an optional thinking mode.

pick it if

  • want Apache 2.0 open weights you can also self-host
  • need image input and a thinking mode at a low flat rate
  • write long outputs, since output costs the same as input

skip it if

  • you need audio input; Google limits it to smaller Gemma 4 sizes
  • you rely on cached-prompt discounts to cut repeat-context cost

GPT 5.6 Luna

GPT 5.6 Luna was the lowest-priced tier of OpenAI's GPT-5.6 family, below Terra and Sol, which OpenAI says roughly matches the earlier nano tier. it retired on 2026-10-06, replaced by the cheaper GPT-6 Luna.

pick it if

  • you needed the cheapest GPT-5.6 tier for high-volume text and image work
  • you reused long prompts, since cached reads billed at a tenth of input
  • you wanted adjustable reasoning effort, from none up to max

skip it if

  • GPT-6 Luna is available: same window, half the input price

faq

which is cheaper, Gemma 4 31B or GPT 5.6 Luna?

Gemma 4 31B lists the lower output price; input prices match. on a coding-agent session (2M tokens in, 60,000 out) Gemma 4 31B costs $0.41 and GPT 5.6 Luna $0.47.

which has the larger context window?

GPT 5.6 Luna offers the larger context window. Gemma 4 31B: 128k tokens; GPT 5.6 Luna: 1M tokens.

try either model

switch between models inside the same octomind session.

prices from the catalog of 2026-11-18