BenchLeader

Gemma 3 27B

Quality#268 of 333
39.9Index
Bottom quarter
−10.1 on the median
Price#22 of 249
$0.150per 1M tokens
Among the cheapest 10%
0.2× the median
Speed#196 of 215
34tokens/s
Bottom quarter
0.4× the median

8 cheaper ranked models score higher: GPT-5 nano (55.9, $0.138), Ling 3.0 Flash (51.8, $0.115), Qwen3.5-4B (49.4, $0.060) and 5 more.

As of 8 Nov 2026, Gemma 3 27B by Google scores 39.9 on the BenchLeader Index (593 ranked configurations), with its best results in human preference (53) and weakest in long context (27). API pricing: $0.100 per 1M input tokens and $0.300 per 1M output tokens ($0.150 blended at 3:1), list price on 8 Nov 2026. It streams about 34 tokens/s (Artificial Analysis). Released 12 Mar 2025.

Key facts (as of 8 Nov 2026)

  • Gemma 3 27B API pricing: $0.100 per 1M input tokens and $0.300 per 1M output tokens ($0.150 blended at 3:1) (list price, 8 Nov 2026).
  • Gemma 3 27B streams about 34 tokens/s (Artificial Analysis, 8 Nov 2026).
  • Gemma 3 27B was released on 12 Mar 2025 (Google).

In detail. Gemma 3 27B is a Google open-weights model, released 12 Mar 2025. Its best configuration ranks #522 of 593 on the BenchLeader Index at 39.9 ±5.0, in the lower half. The ± is the point: 168 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in human preference (53) and lowest in long context (27). At $0.150 per million tokens blended it is among the cheapest fifth of ranked models. Output speed of 34 tokens per second puts it in the slowest quarter, with a first answer in 3.6 s. Last measured 6 Nov 2026.

Blended price
$0.150/M
$0.100 in · $0.300 out
via models.dev
Output speed
34 tok/s
measured by Artificial Analysis
First answer
3.62 s
first token 0.71 s
Context
131k
via LiteLLM price table
Full answer
69 s
median, reasoning included
Cost per task
–
Tokens per task
–
Released
12 Mar 2025
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Overall index40
  2. Reasoning39
  3. Coding28
  4. Agents & tools35
  5. Maths47
  6. Instruction following36
  7. Human preference53
  8. Multimodal40
  9. Long context27

Get an email when Google ships a new model

And once when it gets ranked, plus the weekly review on Mondays.

Choose providers and emails

Versions

Google has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.

ModelReleasedIndexRank
Gemma 3 27Bthis page12 Mar 202539.9#522
Gemma 2 27B24 Jun 202440.8#509

How its index has moved

8 Sep 2026 to 28 Oct 2026
344250

The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.

Benchmark results

See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Instruction following

Benchmarknot statedSourceTrend
IFBench31.8%#353Artificial Analysis
Arena Instruction Followingnot in index1343#201Arena (LMArena)

Human preference

Benchmarknot statedSourceTrend
Arena Text1365#191Arena (LMArena)
Arena Creative Writingnot in index1348#176Arena (LMArena)
Arena Multi-turnnot in index1358#193Arena (LMArena)
Arena Longer Queriesnot in index1365#197Arena (LMArena)
Lech Mazur Creative Writingnot in index79.9%#18Lech Mazur

Multimodal

Benchmarknot statedSourceTrend
Arena Vision1165#110Arena (LMArena)
MMMU-Pro48.0%#260Artificial Analysis
MMMU (validation)64.9%#28MMMU
GeoBenchnot in index52.0%#20GeoBench

Long context

Benchmarknot statedSourceTrend
AA-LCR v1.17.3%#492Artificial Analysis

Composite

Benchmarknot statedSourceTrend
Epoch Capabilities Indexnot in index127.7#147Epoch AI Benchmarking Hub
AA Intelligence Index v4.3.2not in index4.8#626Artificial Analysis
AA Progress Index v1not in index160.1#530Artificial Analysis

API pricing: where to run it

Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.

ProviderSpeedFirst tokenInput $/MOutput $/MContextQuantisation
Nebius Token Factory48 tok/s1.17 s$0.100$0.300110kfp8
Parasail41 tok/s0.76 s$0.080$0.450131kfp8
DeepInfra37 tok/s1.12 s$0.080$0.160131kfp8
NextBit34 tok/s0.51 s$0.080$0.350131kfp8

Price history

The price above, per 1M tokens, every day it changed.

$0.00$0.083$0.165$0.247$0.330Sep 26Oct 26Nov 26
input outputnow $0.100 in · $0.300 out

What a task costs

Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.100 per 1M. Reasoning tokens are not modelled.

WorkloadTokens in / outCostWith cachingTime
Chat reply400 / 300$0.0001$0.000112.5 s
Summarize a 30-page report12,000 / 600$0.0014$0.001421.4 s
Code edit6,000 / 1,500$0.0011$0.001148.1 s
Agentic coding session60,000 / 4,000$0.0072$0.00722.0 min
Structured extraction2,000 / 200$0.0003$0.00039.5 s

More from Google

New this month

See also

Embed or cite

Show where Gemma 3 27B ranks on a model card or README. The badge redraws itself with every daily update, carries the data date and links back to this page. To see ranks like this on OpenRouter and Hugging Face model pages, get the Chrome extension.

[![Gemma 3 27B on BenchLeader: #261 overall](https://www.benchleader.com/badge/gemma-3-27b.svg)](https://www.benchleader.com/models/gemma-3-27b)

Cite this

The ranking changes daily, so a reference carries the data date. APA style:

BenchLeader. (2026, November 8). Gemma 3 27B on the BenchLeader Index (method v1.4) [Data set]. Retrieved November 8, 2026, from https://www.benchleader.com/models/gemma-3-27b

BibTeX:

@misc{benchleader_gemma_3_27b_20261108,
  author       = {{BenchLeader}},
  title        = {{Gemma 3 27B on the BenchLeader Index (method v1.4)}},
  year         = {2026},
  month        = nov,
  howpublished = {\url{https://www.benchleader.com/models/gemma-3-27b}},
  note         = {Data as of 2026-11-08}
}

Every badge style, the embeddable top-10 widget and how to cite the Index

Questions people ask

What are Gemma 3 27B's benchmark scores?
As of 8 Nov 2026, Gemma 3 27B scores 39.9 ±5.0 on the BenchLeader Index, strongest in human preference (53) and weakest in long context (27). That rests on independent benchmark results only, and every score on this page links to its publisher.
How does Gemma 3 27B rank?
Gemma 3 27B is #522 of 593 ranked configurations on 8 Nov 2026. 168 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 69.4.
How much does Gemma 3 27B cost per 1M tokens?
As of 8 Nov 2026, Gemma 3 27B costs $0.100 per 1M input tokens and $0.300 per 1M output tokens, $0.150 blended (three input tokens per output token). The list price last changed on 11 Sep 2026.
When was Gemma 3 27B released?
Google released Gemma 3 27B on 12 Mar 2025.

Data as of 8 Nov 2026. Compare with another model.

Cite as: BenchLeader, “Gemma 3 27B: benchmarks, pricing, speed and rank”, data as of 8 Nov 2026, https://www.benchleader.com/models/gemma-3-27b.