Gemma 3 Models by Google DeepMind
19 models2024–2026Up to 131k ctxFrom $0.02/1M input
About
Gemma 3 is a family of 19 AI models by Google DeepMind, released between 2024 and 2026.
Current Variants
Use-when guidance is derived from seed capabilities, context, release, and replacement fields.
19 in view
Together AI - Gemma 3n-e4BCurrent
Use when the workload needs 8k context, 4B parameters, and tool use.
2026-038k context4B parameterstool use
Gemma 3 12BCurrent
Use when the workload needs 33k context, 12B parameters, and structured outputs.
2026-0133k context12B parametersstructured outputs
Gemma 3 27B PTCurrent
Use when the workload needs 128k context, 27B parameters, and structured outputs.
2026-01128k context27B parametersstructured outputs
Gemma 3 4B ITCurrent
Use when the workload needs 128k context, 4B parameters, and structured outputs.
2026-01128k context4B parametersstructured outputs
Gemma 3n 4B (free)Current
Use when the workload needs 8k context and structured outputs.
2025-048k contextstructured outputs
Gemma 3nCurrent
Use when the workload needs 32k context and structured outputs.
2025-0332k contextstructured outputs
Gemma 3 27BCurrent
Use when the workload needs 131k context, 27B parameters, and structured outputs.
2025-03131k context27B parametersstructured outputs
Gemma 3 12B InstructCurrent
Use when the workload needs 128k context and 12B parameters.
2025-01128k context12B parameters
Gemma 3 1B InstructCurrent
Use when the workload needs 32k context and 1B parameters.
2025-0132k context1B parameters
Gemma 3 27B InstructCurrent
Use when the workload needs 128k context and 27B parameters.
2025-01128k context27B parameters
Gemma 3 4B InstructCurrent
Use when the workload needs 128k context and 4B parameters.
2025-01128k context4B parameters
ShieldGemma 2Current
Use when the workload needs safety, 4B parameters, and tool use.
2024-09safety4B parameterstool use
MedGemmaCurrent
Use when the workload needs 4B parameters, tool use, and function calling.
2024-074B parameterstool usefunction calling
MedSigLIPCurrent
Use when the workload needs 400M parameters, tool use, and function calling.
2024-07400M parameterstool usefunction calling
TxGemmaCurrent
Use when the workload needs 2B parameters, tool use, and function calling.
2024-062B parameterstool usefunction calling
T5GemmaCurrent
Use when the workload needs 2B parameters, tool use, and function calling.
2024-042B parameterstool usefunction calling
PaliGemmaCurrent
Use when the workload needs 3B parameters, tool use, and function calling.
2024-033B parameterstool usefunction calling
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Together AI - Gemma 3n-e4B | Use when the workload needs 8k context, 4B parameters, and tool use. | 2026-03 | 8k context4B parameterstool use | Current |
| Gemma 3 12B | Use when the workload needs 33k context, 12B parameters, and structured outputs. | 2026-01 | 33k context12B parametersstructured outputs | Current |
| Gemma 3 27B PT | Use when the workload needs 128k context, 27B parameters, and structured outputs. | 2026-01 | 128k context27B parametersstructured outputs | Current |
| Gemma 3 4B IT | Use when the workload needs 128k context, 4B parameters, and structured outputs. | 2026-01 | 128k context4B parametersstructured outputs | Current |
| Gemma 3n 2B (free) | Use when the workload needs 8k context. | 2025-04 | 8k context | Current |
| Gemma 3n 4B (free) | Use when the workload needs 8k context and structured outputs. | 2025-04 | 8k contextstructured outputs | Current |
| Gemma 3 | Use when the workload needs structured outputs. | 2025-03 | structured outputs | Current |
| Gemma 3n | Use when the workload needs 32k context and structured outputs. | 2025-03 | 32k contextstructured outputs | Current |
| Gemma 3 27B | Use when the workload needs 131k context, 27B parameters, and structured outputs. | 2025-03 | 131k context27B parametersstructured outputs | Current |
| Gemma 3 12B Instruct | Use when the workload needs 128k context and 12B parameters. | 2025-01 | 128k context12B parameters | Current |
| Gemma 3 1B Instruct | Use when the workload needs 32k context and 1B parameters. | 2025-01 | 32k context1B parameters | Current |
| Gemma 3 27B Instruct | Use when the workload needs 128k context and 27B parameters. | 2025-01 | 128k context27B parameters | Current |
| Gemma 3 4B Instruct | Use when the workload needs 128k context and 4B parameters. | 2025-01 | 128k context4B parameters | Current |
| ShieldGemma 2 | Use when the workload needs safety, 4B parameters, and tool use. | 2024-09 | safety4B parameterstool use | Current |
| MedGemma | Use when the workload needs 4B parameters, tool use, and function calling. | 2024-07 | 4B parameterstool usefunction calling | Current |
| MedSigLIP | Use when the workload needs 400M parameters, tool use, and function calling. | 2024-07 | 400M parameterstool usefunction calling | Current |
| TxGemma | Use when the workload needs 2B parameters, tool use, and function calling. | 2024-06 | 2B parameterstool usefunction calling | Current |
| T5Gemma | Use when the workload needs 2B parameters, tool use, and function calling. | 2024-04 | 2B parameterstool usefunction calling | Current |
| PaliGemma | Use when the workload needs 3B parameters, tool use, and function calling. | 2024-03 | 3B parameterstool usefunction calling | Current |
Release Timeline
10 release groups2026-03
1 current
Together AI - Gemma 3n-e4B
Current8k context4B parameterstool use
2026-01
3 current
Gemma 3 12B
Current33k context12B parametersstructured outputs
Gemma 3 27B PT
Current128k context27B parametersstructured outputs
Gemma 3 4B IT
Current128k context4B parametersstructured outputs
2025-04
2 current
Gemma 3n 2B (free)
Current8k context
Gemma 3n 4B (free)
Current8k contextstructured outputs
2025-03
3 current
Gemma 3
Currentstructured outputs
Gemma 3 27B
Current131k context27B parametersstructured outputs
Gemma 3n
Current32k contextstructured outputs
2025-01
4 current
Gemma 3 12B Instruct
Current128k context12B parameters
Gemma 3 1B Instruct
Current32k context1B parameters
Gemma 3 27B Instruct
Current128k context27B parameters
Gemma 3 4B Instruct
Current128k context4B parameters
2024-09
1 current
ShieldGemma 2
Currentsafety4B parameterstool use
2024-07
2 current
2024-06
1 current
TxGemma
Current2B parameterstool usefunction calling
Specifications(19 models)
| Model | Released | Context | Parameters | Vision | Multimodal | Fn Calling | Tool Use | Structured Outputs |
|---|---|---|---|---|---|---|---|---|
| Together AI - Gemma 3n-e4B | 2026-03 | 8k | 4B | No | No | Yes | Yes | Yes |
| Gemma 3 12B | 2026-01 | 33k | 12B | No | No | No | No | Yes |
| Gemma 3 27B PT | 2026-01 | 128k | 27B | No | No | No | No | Yes |
| Gemma 3 4B IT | 2026-01 | 128k | 4B | No | No | No | No | Yes |
| Gemma 3n 2B (free) | 2025-04 | 8k | 5B (2B effective active) | No | No | No | No | No |
| Gemma 3n 4B (free) | 2025-04 | 8k | 8B (4B effective active) | No | No | No | No | Yes |
| Gemma 3 | 2025-03 | — | — | No | No | No | No | Yes |
| Gemma 3n | 2025-03 | 32k | — | No | No | No | No | Yes |
| Gemma 3 27B | 2025-03 | 131k | 27B | No | No | No | No | Yes |
| Gemma 3 12B Instruct | 2025-01 | 128k | 12B | No | No | No | No | No |
| Gemma 3 1B Instruct | 2025-01 | 32k | 1B | No | No | No | No | No |
| Gemma 3 27B Instruct | 2025-01 | 128k | 27B | No | No | No | No | No |
| Gemma 3 4B Instruct | 2025-01 | 128k | 4B | No | No | No | No | No |
| ShieldGemma 2 | 2024-09 | — | 4B | Yes | Yes | Yes | Yes | Yes |
| MedGemma | 2024-07 | — | 4B | Yes | Yes | Yes | Yes | Yes |
| MedSigLIP | 2024-07 | — | 400M | Yes | Yes | Yes | Yes | Yes |
| TxGemma | 2024-06 | — | 2B | No | No | Yes | Yes | Yes |
| T5Gemma | 2024-04 | — | 2B | No | No | Yes | Yes | Yes |
| PaliGemma | 2024-03 | — | 3B | Yes | Yes | Yes | Yes | Yes |
Available From(9 providers)
Pricing
Frequently Asked Questions
- What is Gemma 3 used for?
- Gemma 3 is used for safety, vision and multimodal work, and agent workflows and tool use. The family description and listed model capabilities point to those workloads as the best fit.
- How does Gemma 3 compare to Gemma 4?
- Gemma 3 by Google DeepMind is strongest where you need safety, while Gemma 4 by Google DeepMind is the closest related family to check for multimodal. Gemma 3 has 19 listed variants and reaches up to 131k context, while Gemma 4 reaches up to 256k context, so compare the specs and pricing tables before choosing a production model.
- Which Gemma 3 model should I use?
- For the lowest listed input price, start with Together AI - Gemma 3n-e4B through Together AI at $0.02/1M input tokens. For the most capable/latest local choice, evaluate ShieldGemma 2 with tool use, function calling, structured outputs, and multimodal inputs.
Related Model Families
Models(19)
Together AI - Gemma 3n-e4B
2026-038k4B1 provider
Open Source
Gemma 3 12B
2026-0133k12B5 providers
Gemma 3 27B PT
2026-01128k27B1 provider
Gemma 3 4B IT
2026-01128k4B3 providers
Gemma 3n 2B (free)
2025-048k5B (2B effective active)1 provider
Gemma 3n 4B (free)
2025-048k8B (4B effective active)3 providers
Gemma 3
2025-033 providers
Open Source
Gemma 3n
2025-0332k2 providers
Open Source
Gemma 3 27B
2025-03131k27B5 providers
Gemma 3 12B Instruct
2025-01128k12B1 provider
Open Source
Gemma 3 1B Instruct
2025-0132k1B2 providers
Open Source
Gemma 3 27B Instruct
2025-01128k27B1 provider
Open Source
Gemma 3 4B Instruct
2025-01128k4B1 provider
Open Source
ShieldGemma 2
2024-094B1 provider
Multimodal
MedGemma
2024-074B1 provider
Multimodal
MedSigLIP
2024-07400M1 provider
Multimodal
TxGemma
2024-062B1 provider
T5Gemma
2024-042B1 provider
PaliGemma
2024-033B1 provider
Multimodal