Gemma 3 27B IT
Summary
Gemma 3 27B IT is an open-weight language model from Google DeepMind, released on 12 Mar 2025. It accepts text and images and generates text, with a 131K-token context window and up to 131K output tokens per response. As of 18 Nov 2026, the lowest listed API price is $0.08 per million input tokens and $0.16 per million output tokens (DeepInfra), across 5 providers we track.
Open Gemma instruction model for efficient chat and self-hosted deployments
Last verified 18 Nov 2026. Every figure on this page links to its source.
Specifications
| Developer | Google DeepMind |
|---|---|
| Release date | 12 Mar 2025 [models.dev] |
| Status | Generally available [models.dev] |
| Weights | Open weights [Hugging Face] |
| License | gemma [Hugging Face] |
| Parameters | 27.4B [Hugging Face] |
| Context window | 131K (131,072 tokens) [models.dev] |
| Max output | 131K tokens [models.dev] |
| Knowledge cutoff | Aug 2024 [models.dev] |
| Input | Text and images [models.dev] |
| Output | Text [models.dev] |
| Reasoning mode | No [models.dev] |
| Tool calling | Yes [models.dev] |
| Structured output | Yes [models.dev] |
| Hugging Face | google/gemma-3-27b-it [Hugging Face] |
Gemma 3 27B IT API pricing by provider
USD per million tokens. Sorted by input price.
| Provider | Input | Output | Cached input | Context | Source |
|---|---|---|---|---|---|
| DeepInfra google/gemma-3-27b-it | $0.08 | $0.16 | – | 131K | models.dev, 18 Nov 2026 |
| Hugging Face Inference Providers google/gemma-3-27b-it | $0.08 | $0.16 | – | 131K | models.dev, 18 Nov 2026 |
| OpenRouter google/gemma-3-27b-it | $0.08 | $0.35 | $0.04 | 131K | OpenRouter, 18 Nov 2026 |
| Nebius AI Studio google/gemma-3-27b-it | $0.10 | $0.30 | $0.10 | 110K | models.dev, 18 Nov 2026 |
| Novita AI google/gemma-3-27b-it | $0.12 | $0.20 | – | 98K | models.dev, 18 Nov 2026 |
Running Gemma 3 27B IT locally
Estimated memory, from the parameter count on Hugging Face. Includes about 0.8 GB of runtime overhead.
| Quantization | Weights | Total, 8K context | Total, 32K context |
|---|---|---|---|
| Q4_K_M (4-bit) | 17 GB | 17 GB | – |
| Q8_0 (8-bit) | 29 GB | 30 GB | – |
| FP16 / BF16 (16-bit) | 55 GB | 56 GB | – |
At 4-bit with an 8K conversation, Gemma 3 27B IT needs about 17 GB, so it fits on a graphics card with 24 GB or more, such as the NVIDIA GeForce RTX 3090 24 GB. The model's architecture is not public, so the conversation cache is not included. Check your own hardware, or see Best GPU for local LLMs on our sister site AIPCs.
Frequently asked questions
What is the context window of Gemma 3 27B IT?
Gemma 3 27B IT has a context window of 131,072 tokens (131K) and can generate up to 131,072 tokens in a single response, according to models.dev.
How much does the Gemma 3 27B IT API cost?
Gemma 3 27B IT is listed from $0.08 per million input tokens and $0.16 per million output tokens (DeepInfra). Prices last verified 18 Nov 2026.
Is Gemma 3 27B IT open source?
Gemma 3 27B IT is an open-weight model: its weights are published on Hugging Face as google/gemma-3-27b-it under the gemma license. Check the license terms before commercial use.
When was Gemma 3 27B IT released?
Google DeepMind released Gemma 3 27B IT on 12 Mar 2025, according to models.dev.
Which providers offer Gemma 3 27B IT?
We track Gemma 3 27B IT on 5 providers: DeepInfra, Hugging Face Inference Providers, OpenRouter, Nebius AI Studio, Novita AI.
How much VRAM does Gemma 3 27B IT need?
About 17 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 30 GB at 8-bit. The 4-bit model fits on a graphics card with 24 GB or more, such as the NVIDIA GeForce RTX 3090 24 GB. These are estimates from the model's published parameter count.
Similar models from other labs
Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.
| Model | Lab | Released | Context | Input $/1M | Output $/1M | Weights |
|---|---|---|---|---|---|---|
| Command R7B Arabic | Cohere | 27 Feb 2025 | 128K | $0.04 | $0.15 | Open |
| Mistral Small 3.1 24B | Mistral AI | 17 Mar 2025 | 128K | $0.10 | $0.30 | Open |
| Llama-4-Scout-17B-16E-Instruct-FP8 | Meta | 5 Apr 2025 | 128K | $0.10 | $0.30 | Open |
| Command R7B | Cohere | 13 Dec 2024 | 128K | $0.04 | $0.15 | Open |
| Qwen2.5-Coder-32B-Instruct | Alibaba (Qwen) | 12 Nov 2024 | 131K | $0.06 | $0.20 | Open |
| Llama Guard 4 12B | Meta | 29 Apr 2025 | 164K | $0.18 | $0.18 | Open |
Change history
- Context window changed from 98K to 131K
- Tool calling changed from no to yes
- Max output changed from 16K to 131K
- Max output changed from 8K to 16K
- Tool calling changed from yes to no
- Name changed from Gemma-3-27b-it to Gemma 3 27B IT
- Context window changed from 110K to 98K
- Max output changed from 131K to 8K
- Context window changed from 131K to 110K
- Name changed from Gemma 3 27B IT to Gemma-3-27b-it
- OpenRouter price changed: input $0.08 → $0.08, output $0.45 → $0.35 per 1M tokens
- Max output changed from 118K to 131K
- Name changed from Gemma-3-27b-it to Gemma 3 27B IT
- Context window changed from 110K to 131K
- Max output changed from 8K to 118K
- Name changed from Gemma 3 27B IT to Gemma-3-27b-it
- Max output changed from 131K to 8K
- Context window changed from 131K to 110K
- OpenRouter price changed: input $0.08 → $0.08, output $0.35 → $0.45 per 1M tokens
- Context window changed from 110K to 131K
- Max output changed from 8K to 131K
- Name changed from Gemma-3-27b-it to Gemma 3 27B IT
- Context window changed from 131K to 110K
- Max output changed from 118K to 8K
- Name changed from Gemma 3 27B IT to Gemma-3-27b-it
Sources
- https://openrouter.ai/google/gemma-3-27b-it
- https://github.com/sst/models.dev/blob/dev/providers/huggingface/models/google/gemma-3-27b-it.toml
- https://huggingface.co/google/gemma-3-27b-it
- https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/google/gemma-3-27b-it.toml
- https://github.com/sst/models.dev/blob/dev/providers/nebius/models/google/gemma-3-27b-it.toml
- https://github.com/sst/models.dev/blob/dev/providers/novita-ai/models/google/gemma-3-27b-it.toml