Gemma 3 12B
Google · released Mar 1, 2025 · google/gemma-3-12b-it
- Type
- Open weightsGemma Terms of Use
- Params
- 12.2B
- Context
- 131K
about 98K words of context · download allowed, licence restricts use
Our take
Written Aug 3, 2026Gemma 3 is a compact vision-language model from Google that accepts text and images and can handle up to 131,072 tokens in a single request. Its weights can be downloaded, though under terms that restrict some commercial uses, and it sits at a cheap entry point for multimodal inference.
Pick this for budget image-and-text workloads where low cost matters more than licensing flexibility, or for long-context tasks needing six-figure token limits. Use it if you are already inside the Google or Gemma ecosystem. Skip it if you need to redistribute the weights, use it in ways the Gemma Terms of Use restrict, or if you need verified speed guarantees from your chosen host.
The case for it
- Very cheap entry point for multimodal inference: the cheapest tracked offer undercuts the next one by a third on output cost.
- 131,072-token request limit — no smaller-context variant in our data for this parameter class.
- Consistent leaderboard presence across six task categories, with scores stable to within a point between July and August evaluations.
The case against it
- Gemma Terms of Use restrict redistribution and certain commercial uses — not as permissive as Apache 2.0.
- Throughput is unverified for three of four tracked hosts; only one lists a measured speed.
- Dense 12.2-billion-parameter architecture with no disclosed active-parameter count, so no efficiency gains from a mixture-of-experts design.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)107th of 143 · 1342
CodingWriting and fixing code on its own
Arena Coding127th of 143 · 1316.2
Arena Coding is the only board that has scored it for this.
AgenticPlanning, calling tools, staying on task
Nobody we watch has scored Gemma 3 12B for this. We would take the rating from Arena Agent (IPS).
WritingWe do not rate this
Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage. So we show where Gemma 3 12B placed and give it no mark out of five.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.
Every published score for this model6 scoresEvery figure we hold, from 6 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Can you run it yourself?
- Fits in memory
- weights load entirely on the card
- Spills to system RAM
- some weights offload; much slower
- Too large
- will not load even with offload
- est
- size is calculated; the verdict could change by 10%
Comfortable fit
GeForce RTX 4090 · 24 GB
Room to spare. 13.6 GB spare means a 10% error in the size would not change the answer.
GeForce RTX 5090 · 32 GB
Room to spare. 21.6 GB spare means a 10% error in the size would not change the answer.
Apple M1 (8-core GPU) · 16 GB
Room to spare. 2.8 GB spare means a 10% error in the size would not change the answer.
Memory use by level
Against a 24 GB card.
All 3 sizes here are calculated, not measured. We hold no measured file for this model, so each size comes from the parameter count and every verdict above inherits that uncertainty.
Check against your own machine → · All 71 devices, with every size →
Or rent it from someone else
Cheapest of 4 live listings. Picked at the widest standard context we hold, within one quantisation slice, so the numbers beside it are a price one host actually charges.
- per 1M tokens
- $0.050 in / $0.10 out
- Context served
- 131K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| Novita AI | $0.050 / $0.10 | 131K | not measured | Unknown | Unknown | Unknown |
| OpenRouter | $0.050 / $0.15 | 131K | not measured | Unknown | Unknown | Unknown |
| DeepInfrabf16 | $0.050 / $0.15 | 131K | 45 tok/s | No | No | Confirmed |
| DeepInfrabfloat16 | $0.050 / $0.15 | 131K | not measured | Unknown | Unknown | Unknown |
OpenRouter
DeepInfraAcross the 4 listings we hold: 1 say they do not train on prompts, 0 say they do and 3 do not say. 1 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| Novita AI | |||
| OpenRouter | ✓ | ✓ | ✓ |
| DeepInfrabf16 | ✓ | ✓ | ✓ |
| DeepInfrabfloat16 |
Tool calling: 2 of 4 listings say yes, 2 publish no parameter list. JSON output: 2 of 4 listings say yes, 2 publish no parameter list. Strict schema: 2 of 4 listings say yes, 2 publish no parameter list.
Models people weigh against Gemma 3 12B
When we formed this view
Dates behind this page
Prices last checked 6h ago
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- 2 of 4 listings publish no parameter list, so what their API accepts is unknown to us.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 3 of 4 listings do not say whether they train on prompts.
- We hold no cached-input rate for any of its listings.
Licence and identifiers
What the licence allowsGemma Terms of Use, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
Gemma Terms of Use
Commercial use allowed, but Google's prohibited-use policy applies and can be updated over time — terms are less static than Apache/MIT.
Identifiers
- Hugging Face
- google/gemma-3-12b-it
- Modality record
- text+image->text
- Catalogue slug
- google-gemma-3-12b