Gemma 3 27B IT

Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities, including structured outputs and function calling. Gemma 3 27B is Google's latest open source model, successor to Gemma 2

google/gemma-3-27b-it

Organization: Google
Family: gemma
Providers: 17
Context: 131,072
Output limit: 131,072
Knowledge: 2024-08
Release: 2025-03-12
Updated: —
Weights: Open
Input: text, image/png, image/jpeg, image/webp, image/heic, image/heif
Output: text
Capabilities: —

Providers

Compare every cataloged offer for this model, including its provider, route, token limits, price, and capabilities.

25 offers · USD per million tokens

Provider	Model ID	Context	Output	Input / output · 1M	Reasoning	Tools	Structured	Details
AI/ML API	google/gemma-3-27b-it	128,000	117,964	— / —	—	—	—	View
Amazon Bedrock	google.gemma-3-27b-it	131,072	8,192	— / —	—	—	—	View
Cortecs	gemma-3-27b-it	131,000	110,000	— / —	—	—	—	View
Deep Infra	google/gemma-3-27b-it	131,072	131,072	— / —	—	—	—	View
Eden AI	deepinfra/google/gemma-3-27b-it	131,072	131,072	— / —	—	—	—	View
Eden AI	amazon/google.gemma-3-27b-it	128,000	131,072	— / —	—	—	—	View
Eden AI	nebius/google/gemma-3-27b-it	110,000	131,072	— / —	—	—	—	View
Eden AI	scaleway/gemma-3-27b-it	40,000	131,072	— / —	—	—	—	View
Eden AI	amazon/google.gemma-3-27b-it@us	128,000	131,072	— / —	—	—	—	View
EmpirioLabs AI	gemma-3-27b	128,000	8,192	— / —	—	—	—	View
GreenPT	gemma-3-27b-it	40,000	8,192	— / —	—	—	—	View
Hugging Face	google/gemma-3-27b-it	131,072	131,072	— / —	—	—	—	View
Kilo Gateway	google/gemma-3-27b-it	131,072	117,964	— / —	—	—	—	View
Merge Gateway	google/gemma-3-27b-it	131,072	32,768	— / —	—	—	—	View
NanoGPT	unsloth/gemma-3-27b-it	128,000	96,000	— / —	—	—	—	View
Nebius Token Factory	google/gemma-3-27b-it	110,000	8,192	— / —	—	—	—	View
Novita AI	google/gemma-3-27b-it	98,304	16,384	— / —	—	—	—	View
DeepInfravia OpenRouter	google/gemma-3-27b-it fp8	131,072	16,384	$0.08 / $0.16	No	Yes	Yes	View
Nebiusvia OpenRouter	google/gemma-3-27b-it fp8	110,000	99,000	$0.10 / $0.30	No	No	Yes	View
NextBitvia OpenRouter	google/gemma-3-27b-it fp8	131,072	117,964	$0.08 / $0.30	No	No	Yes	View
Novitavia OpenRouter	google/gemma-3-27b-it bf16	98,304	16,384	$0.119 / $0.20	No	No	No	View
Parasailvia OpenRouter	google/gemma-3-27b-it fp8	131,072	117,964	$0.08 / $0.45	No	No	Yes	View
Prime Inference	google/gemma-3-27b-it	131,072	117,964	— / —	—	—	—	View
STACKIT	google/gemma-3-27b-it	37,000	4,096	— / —	—	—	—	View
Venice AI	google-gemma-3-27b-it	198,000	16,384	— / —	—	—	—	View

Prices and limits apply to each offer. A dash means the value is not provided. OpenRouter offers show the hosting provider and routing channel separately.

Pricing

Compare token rates and cache charges across available routes and hosting configurations.

USD per million tokens

Input from$0.08Per 1M input tokens

Output from$0.16Per 1M output tokens

Available offers25Across 17 providers

Lowest listed input and output prices across the offers above; they may belong to different providers. These are base rates, before conditional pricing, discounts or additional fees.

Endpoint price comparisonLatest observed snapshot · 5 lowest-priced configurations

InputOutput

DeepInfradeepinfra/fp8

$0.08$0.16

NextBitnextbit/fp8

$0.08$0.30

Parasailparasail/fp8

$0.08$0.45

Nebiusnebius/fp8

$0.10$0.30

Novitanovita/bf16

$0.119$0.20

google/gemma-3-27b-itObserved Oct 10, 2026, 11:31 AM UTC

Input: $0.08
Output: $0.45
Cache read: $0.04

Hosting prices through OpenRouter All endpoints and cache rates

Endpoint	Input / 1M	Output / 1M	Cache read / 1M	Cache write / 1M
DeepInfradeepinfra/fp8	$0.08	$0.16	—	—
Nebiusnebius/fp8	$0.10	$0.30	—	—
NextBitnextbit/fp8	$0.08	$0.30	$0.04	—
Novitanovita/bf16	$0.119	$0.20	—	—
Parasailparasail/fp8	$0.08	$0.45	$0.04	—

Each row is a hosting configuration. Open an offer in Providers for its full pricing conditions and additional fees. Endpoint observations may differ from the routing catalog's base price.

Performance

Review reported latency and output-throughput percentiles from the latest observed rolling window.

Latest 30-minute observation

Time to first token · milliseconds

Latency distributionRolling 30-minute snapshot · 5 endpoints

P50P75P90P99

DeepInfradeepinfra/fp8

1,524–10,206.5 ms

Nebiusnebius/fp8

773–3,442 ms

NextBitnextbit/fp8

733–2,558.45 ms

Novitanovita/bf16

1,255–7,492.24 ms

Parasailparasail/fp8

633.5–8,044.31 ms

All latency percentiles 5 endpoints

Endpoint	P50	P75	P90	P99
DeepInfradeepinfra/fp8	1,524 ms	2,440.5 ms	4,228 ms	10,206.5 ms
Nebiusnebius/fp8	773 ms	1,072 ms	1,640 ms	3,442 ms
NextBitnextbit/fp8	733 ms	1,193.25 ms	1,518 ms	2,558.45 ms
Novitanovita/bf16	1,255 ms	1,625 ms	2,294.4 ms	7,492.24 ms
Parasailparasail/fp8	633.5 ms	912.25 ms	1,575.5 ms	8,044.31 ms

Output throughput · tokens per second

Throughput distributionRolling 30-minute snapshot · 5 endpoints

P50P75P90P99

DeepInfradeepinfra/fp8

25–50 t/s

Nebiusnebius/fp8

60–87 t/s

NextBitnextbit/fp8

52–64 t/s

Novitanovita/bf16

36–48.08 t/s

Parasailparasail/fp8

38–72 t/s

All throughput percentiles 5 endpoints

Endpoint	P50	P75	P90	P99
DeepInfradeepinfra/fp8	25 t/s	33 t/s	39 t/s	50 t/s
Nebiusnebius/fp8	60 t/s	70 t/s	77 t/s	87 t/s
NextBitnextbit/fp8	52 t/s	57 t/s	60 t/s	64 t/s
Novitanovita/bf16	36 t/s	39 t/s	41 t/s	48.08 t/s
Parasailparasail/fp8	38 t/s	46 t/s	53 t/s	72 t/s

Percentiles describe the observed request distribution. These are snapshots of a rolling window, not a historical time series.

Uptime

Compare successful request rates across the latest 5-minute, 30-minute, and 24-hour observations.

Successful requests by window

Availability by endpointCurrent overlapping windows · 5 endpoints

Endpoint5 minutes30 minutes24 hours

DeepInfradeepinfra/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

Nebiusnebius/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

97.01%

NextBitnextbit/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

99.44%

Novitanovita/bf16Observed Oct 10, 2026, 11:31 AM UTC

100%

95.4%

Parasailparasail/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

99.72%

Reported by OpenRouter at the time of collection; rate-limited requests are excluded. A dash means no measurement was reported. This is not a live availability check.

Benchmarks

Review attributed evaluation results on each source's original scale.

External evaluations

google/gemma-3-27b-itObserved Oct 10, 2026, 11:31 AM UTC

Benchmark profileArtificial Analysis · shared source scale from 0 to 20

Coding

10.1

Agentic

0.1

Intelligence

4.9

Sources: Artificial Analysis and Design Arena, via OpenRouter. Scores retain their original scales and are not OpenAI Suite ratings.

Detailed evaluations

Evaluation	Configuration	Source	Results	Details
Composite indices	Gemma 3 27B Instructgoogle/gemma-3-27b-it	Artificial Analysis	Intelligence 4.9 · Coding 10.1 · Agentic 0.1	Details

Sources: Artificial Analysis, Design Arena and OpenRouter. Evaluation configurations and scales differ; results are not interchangeable. Updated Oct 24, 2026.

Technical details & API

Explore specifications, supported parameters, reasoning controls, and API identifiers for each OpenRouter route.

Current route specifications

google/gemma-3-27b-itObserved Oct 10, 2026, 11:31 AM UTC

Context: 131,072
Maximum output: 117,964
Tokenizer: Gemini
Input: text, image
Output: text
Moderation: No
Instruction format: gemma
Knowledge cutoff: 2024-08-31
Added to OpenRouter: 2025-03-12

Base URLhttps://openrouter.ai/api/v1

Model IDgoogle/gemma-3-27b-it

Version IDgoogle/gemma-3-27b-it

Weights IDgoogle/gemma-3-27b-it

Supported parameters

frequency_penalty
logit_bias
logprobs
max_tokens
min_p
presence_penalty
repetition_penalty
response_format
seed
stop
structured_outputs
temperature
tool_choice
tools
top_k
top_logprobs
top_p

Frequently asked questions

Answers based on the model metadata and serving offers currently in the catalog.

How much does Gemma 3 27B IT cost?

Listed rates start at $0.08 per million input tokens and $0.16 per million output tokens. Prices vary by provider and configuration; see Pricing for details.

Which providers offer Gemma 3 27B IT?

There are 25 cataloged offers across 17 providers. Compare model IDs, prices and limits in Providers.

What is the context window?

The cataloged model context is 131,072 tokens. Each serving endpoint may apply a different limit.

Does it support tools and structured output?

Tool calling: Not reported. Structured output: Not reported. Support can vary by endpoint.

Resources

Hugging Face ↗

Catalog history · 3 recorded revisions

Metadata observations since this model was first cataloged. These are not model release versions.

Oct 20, 2026 · a858a1a6047c
Sep 14, 2026 · f643abd1d3c3
Sep 12, 2026 · b9daf676d37d