Gemma 3 27B

Released
2025-03-12
Last refreshed
2026-10-26
Status
Researched 20d ago
Open weightsCommercial use: restrictedCodingRAGLong contextClassificationJSON / Tool use
Google DeepMind releases · 54 in the last 12 months · this family litChangelog →
Specifications
Family
Gemma 3
Released
2025-03-12
Context
131K
Parameters
27B
Architecture
Decoder Only
Knowledge cutoff
2024-08
Specialization
general
Openness
Open weights
License
GemmaRestricted
Weights
Available
Code
Unknown
Created by

Pioneering artificial intelligence research.

London, United Kingdom
Founded 2014
Website
Pricing
Output / 1M
$0.160
Input / 1M
$0.080

Cheapest of 6 routes · GCP Vertex AI

About

Gemma 3 27B is Google DeepMind's Gemma 3 instruction-tuned model with openly available weights and a 131K-token context window.

Top use-case fit: coding, agents, and build tasks

Coding

Q/$ C

RAG

Q/$ B

Long context

Q/$ B

Providers(6)

Compare all 6

Compare API pricing across 6 tracked provider routes for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MRoute
GCP Vertex AI$0.080$0.160
Serverless
Novita AI$0.119$0.200
Serverless
OpenRouter$0.119$0.200
Serverless
AWS Bedrock$0.500$0.500
Serverless
Scheduled retirement: March 30, 2027
Fireworks AI$0.900$0.900
Serverless
NVIDIA NIM--
ServerlessPartial

Available via routers & gateways(15)

Capabilities

Structured Outputs

Benchmark peer barsfor Coding

Benchmark scores(5)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionEvaluationSource
4.92026-04Observed 2026-04-14—
29.5v4Observed 2026-10-26
Configuration: Gemma-3-27b-it (Prompt)
Harness: BFCL V4 official evaluation (gorilla berkeley-function-call-leaderboard), Prompt mode
Evaluator: Gorilla LLM team (UC Berkeley)
Confidence: confirmed
Notes: BFCL V4 leaderboard (Last Updated 2026-04-12), entry 'Gemma-3-27b-it (Prompt)': rank 69, Overall Acc 29.47%; Multi Turn 10.75%, Web Search 0.00%, Memory 13.55%, Non-Live AST 87.17%, Live 74.54%, Irrelevance 73.67%. Retrieved 2026-10-26.
66.0RULER (128K) (accuracy)Observed 2026-10-26—
1160.0vision_style_control overall 2026-10-19 (style control, default view)Observed 2026-10-26
Configuration: Gemma 3 27B IT
Harness: LMArena blind pairwise human votes, Bradley-Terry
Evaluator: LMArena
Confidence: confirmed
Notes: LMArena vision style control leaderboard, overall category, published 2026-10-19 (HF lmarena-ai/leaderboard-dataset @9761a3ce, latest split; live page https://lmarena.ai/leaderboard/vision showed the same 2026-10-19 snapshot). Board entry 'Gemma 3 27B IT': rank 113, Arena Score 1159.9 (95% CI 1151.8-1168.0), 16982 votes. Versioned arena label mapped to catalog slug gemma-3-27b-it (label date equals the model's release checkpoint). Retrieved 2026-10-26.
Source

Source: LMArena (lmarena.ai), CC BY 4.0

1345.0text creative_writing 2026-10-22Observed 2026-10-22
Configuration: Gemma 3 27B IT
Harness: LMArena blind pairwise human votes, Bradley-Terry
Evaluator: LMArena
Confidence: confirmed
Notes: LMArena Text arena, Creative Writing category (no style control), board published 2026-10-22 (HF lmarena-ai/leaderboard-dataset @a467e6b9, text config, latest split). Board entry 'Gemma 3 27B IT' (category creative_writing in text/latest-00000-of-00001.parquet): rank 173 of 416, rating 1345.2 (95% CI 1337.5-1352.8), 6303 votes. Source: LMArena (CC BY 4.0). Retrieved 2026-10-27.
Source

Source: LMArena (lmarena.ai), CC BY 4.0

Migration checks

No linked migration route is available for this model yet.