Gemma 3 4B costs $0.05 per 1M input tokens and $0.10 per 1M output tokens on the standard tier. It holds 131,072 tokens of context.On the atlas: grid B3. 10 other models share the square; the cheapest is gpt-oss-20b at $0.02 per 1M input.
Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
Cite
Cite this page
ModelsAtlas. “Gemma 3 4B”. Retrieved 17 Aug 2026. https://modelsatlas.com/models/google/gemma-3-4b-it
- Input / 1M
$0.05
observed We measured- Source
- Our catalogue
- Retrieved
- 17 Aug 2026
- Output / 1M
$0.10
observed We measured- Source
- Our catalogue
- Retrieved
- 17 Aug 2026
- Context
- tokens
131K
observed We measured- Source
- Our catalogue
- Retrieved
- 17 Aug 2026
- Offers
- 1 offer
Standard tier
Standard tier
≈ 98,000 words
1 host · standard
Hosts and pricing
1 offer · tiers are rows, not pagesSuggest a correction
| Host · tier | Input / 1M | Output / 1M |
|---|---|---|
| Our catalogueStandardchecked 17 Aug 2026 | $0.05observed We measured
| $0.10observed We measured
|
Paying too much? Models near Gemma 3 4B on the atlas.Cheaper alternatives to Gemma 3 4B →
What it costs for your workload
Specifications
Tap src for a sourceSuggest a correction
| Catalogue id | google-gemma-3-4b-it | |
|---|---|---|
| Creator | ||
| Context window | 131,072 tokens | Observed |
| Modalities | Not disclosed | |
| Licence | Not disclosed | |
| Weights | Not disclosed | |
| Variants | One (base) | |
| Record checked | 2026-08-17 |
Capabilities
“Not disclosed” means our records do not say, not that the model lacks it.
- Vision inputNot disclosed
- AudioNot disclosed
- Open weightsNot disclosed
- Free tierNone listed
- Batch tierNone listed
We hold no data yet on tool calling, structured output or benchmark scores, so none are shown.
Cheaper alternatives to Gemma 3 4B
All 15 alternatives →| Model | Grid | Input / 1M | Output / 1M | Context | |
|---|---|---|---|---|---|
| gpt-oss-20bOpenAI | GRIDB3 | $0.02 | $0.09 | 131K | Compare |
| Mistral NemoMistral AI | GRIDB3 | $0.03 | $0.03 | 131K | Compare |
| gpt-oss-120bOpenAI | GRIDB3 | $0.03 | $0.15 | 131K | Compare |
| Llama 3.2 1B InstructMeta | GRIDB2 | $0.03 | $0.20 | 60K | |
| Command R7B (12-2024)Cohere | GRIDB2 | $0.04 | $0.15 | 128K | Compare |
Ranked by distance from grid B3, then price.See all 15 alternatives →
Head-to-head comparisons
Works with
Where Gemma 3 4B connects across the atlas
◆Hosts serving it
- Our catalogue1 offer
◉Tools built on it
- None linked yet. Tell us about one
Questions about Gemma 3 4B
How much does Gemma 3 4B cost?
On the cheapest standard-tier offer in our catalogue, input costs $0.05 per 1M tokens and output costs $0.10 per 1M tokens.
What is the Gemma 3 4B context window?
131,072 tokens, roughly 98,000 English words in one request.
Is there a free tier for Gemma 3 4B?
No free-tier offer is listed for Gemma 3 4B in our catalogue. Its cheapest standard tier is $0.05 per 1M input tokens.
What is the cheapest alternative to Gemma 3 4B?
Among the models nearest it on the atlas, gpt-oss-20b is the cheapest: $0.02 per 1M input tokens, 64% less, with 131K of context.
What does grid B3 mean for Gemma 3 4B?
Grid B3 is input <$0.10 per 1M tokens with context 128–256K. 10 other models share that square: gpt-oss-20b, Mistral Nemo, gpt-oss-120b and 7 more. It describes price and context, not quality.
Sources
Every number on this page, where it came from
- Our catalogue — context window, modalities, licenceObserved · 17 Aug 2026
- Our catalogue — prices and tiersObserved · 17 Aug 2026
- Our calculation — grid reference, monthly cost and percentage comparisonsDerived · 17 Aug 2026
Spotted something wrong? Report a correction · Methodology
Keep going in your AI
Opens with this page as the source. We add nothing else to the prompt.
- How much does Gemma 3 4B cost?
- What is the Gemma 3 4B context window?
- Is there a free tier for Gemma 3 4B?