The Known Good Updated 8 Sep 2026 Subscribe
Google ·Open weights (restricted use) ·Released January 2026·Tracked since 25 Jul 2026

Google: Gemma 3 27B — quality, speed & price analysis

Compare with another model Provider benchmarks Data sources
Quality
#49 of 65
47based on 4 of 7 evaluations
Known Good Index
Speed
#304 of 341
21.0
Output tokens per second
Price
#74 of 450
$0.17
Blended per 1M tokens
In $0.08 · Out $0.45
Cache hit price
#111 of 344
$0.040
per 1M tokens (−50%)
Context
#384 of 487
131k
tokens (~200 pages)
Summary

Google: Gemma 3 27B is in the lowest quarter of the models we track for quality, and inexpensive for its standing, and notably slow. It accepts text, image input, outputs text, and supports a 131k-token context window.

It scores 47 on the Known Good Index, well below the tracked average of 65, from 4 of the seven component evaluations. Measured against the tracked average, its strongest result is MATH Level 5 (74 against 54) and its weakest is Mock AIME 2024–25 (22 against 54).

Pricing is $0.08 per 1M input tokens (below the $2.09 average) and $0.45 per 1M output (below the $9.60 average). Cached input reads at $0.040 per 1M, a 50% discount on its input price. At 21 tokens per second it sits in the slowest quarter of the models we track.

Generated from tracked data on 9 Sep 2026 · how these comparisons are computed →

Technical specifications
ReasoningNo
Input modalitytext, image
Output modalitytext
Context window131,072 tokens
Max output117,964 tokens
WeightsOpen weights · Open weights (restricted use) (commercial use restricted)
Knowledge cutoffAugust 2024
Tool callingYes
Providers5 hosts · compare →
Highlights — where Google: Gemma 3 27B sits
Quality

Known Good Index · Higher is better · Composite of seven published evaluations

97
97
95
93
93
93
92
47
gpt-5.5-pre-release
Gemini 3.1 Pro
Preview
Gemini 3.5 Flash
DeepSeek V4 Pro
0423
GPT-5.5
Qwen3.7 Max
Claude Opus 4.7
Gemma 3 27Bthis #49/65
Speed

Output tokens per second · Higher is better · OpenRouter endpoint telemetry

5,945
3,136
386
350
339
335
318
21
Relace Apply 3
Morph V3 Fast
Grok 4.20
Multi-Agent
Mercury 2.5
Nano Banana 2 Lite
(Gemini 3.1 Flash Lite Image)
Mercury 2.5
Preview
gpt-oss-safeguard-20b
Gemma 3 27Bthis #304/341 · below scale
Price

Blended USD per 1M tokens · Lower is better · Published pricing via OpenRouter

$0.01
$0.02
$0.03
$0.04
$0.04
$0.04
$0.05
$0.17
Ling-2.6-flash
Mistral Nemo
Ling 3.0 Flash
Granite 4.0 Micro
Llama 3 8B Lunaris
Nex-N2-Mini
Solar Pro 4
Gemma 3 27Bthis #74/450 · off scale
Jump to section
Models shown

Quality

How Google: Gemma 3 27B compares against every tracked model

Known Good IndexCodingMathematicsReasoning
Known Good Index

Composite of GPQA Diamond, Mock AIME 2024–25, MATH Level 5, SWE-bench Verified, LiveBench, Humanity's Last Exam and Terminal-Bench, normalised 0–100. Scores are ingested from Epoch AI, LiveBench; we do not run these evaluations.

⇩
97
97
95
93
93
93
92
92
91
91
90
89
89
47
💡 A lightbulb marks a model its provider publishes as a reasoning model
The Known Good

Quality Breakdown

Google: Gemma 3 27B on each evaluation feeding the index, against the tracked average

Evaluation scores

Bar = Google: Gemma 3 27B · grey marker = average across tracked models · scores ingested from Epoch AI, LiveBench

GPQA Diamond48 / 64
Mock AIME 2024–2522 / 54
MATH Level 574 / 54
LiveBench50 / 52
The Known Good

Arena Elo

Human preference rating from blind pairwise votes

Text Arena

Elo from blind pairwise votes, ingested from Arena

1358
Elo · 95% CI ±4
#183
of 408 ranked models
46k
votes counted
1365
style-controlled (#185)

Ingested 8 Sep 2026.

The Known Good

Price & Cost

Published API pricing by token type

BlendedInput vs outputQuality per dollar
Blended Price

USD per 1M tokens at a 3:1 input:output blend. Live from OpenRouter, cross-checked against models.dev. Lower is better. Ranked against the 450 models priced above zero, which is the population the chart draws; excludes 32 published at no cost (free and promotional tiers).

⇩
$0.01
$0.02
$0.03
$0.04
$0.04
$0.04
$0.05
$0.06
$0.06
$0.06
$0.06
$0.06
$0.06
$0.17
The Known Good
We publish blended price rather than cost-per-task. Cost-per-task requires token counts from an evaluation suite we do not run. Blended price is computed from published rates and is fully reproducible — see the pricing methodology.

Context Window

Maximum input tokens accepted

Context windowMax output tokens
Context Window

Maximum input context from provider metadata via models.dev and OpenRouter. Higher is better.

⇩
2.00M
2.00M
2.00M
2.00M
2.00M
1.31M
1.31M
1.31M
1.31M
1.31M
1.31M
1.31M
131k
The Known Good

Speed

Mean output throughput across hosting providers, over successful measurements from the last 48 hours

Output speedBy provider
Output Speed

Mean output tokens per second across all tracked providers, over successful measurements from the last 48 hours. Derived from OpenRouter endpoint telemetry. Higher is better.

⇩
5,945
3,136
386
350
339
335
318
245
179
166
165
163
21
The Known Good

Latency

Time to first token, mean across providers over successful measurements from the last 48 hours

Latency (TTFT)

Seconds to first streamed chunk, mean across tracked providers over successful measurements from the last 48 hours. Derived from OpenRouter endpoint telemetry. Lower is better.

⇩
0.1
0.2
0.2
0.2
0.2
0.3
0.3
0.3
0.3
0.3
0.3
0.3
1.6
The Known Good

Providers

Hosts serving Google: Gemma 3 27B, with their own price and speed

5 providers
Full provider benchmarks →
ProviderInput $/MOutput $/M BlendedSpeedTTFTContext
Phala $0.15 $0.46 $0.23 51 0.8s 262k
Parasail $0.08 $0.45 $0.17 27 0.9s 131k
Nebius $0.10 $0.30 $0.15 18 1.1s 110k
DeepInfra $0.08 $0.16 $0.10 14 2.7s 131k
Novita $0.12 $0.20 $0.14 14 1.4s 98k

Per-host pricing and telemetry from OpenRouter. Speed is output tokens per second; TTFT is time to first token.

The Known Good

History

How price and quality have moved since we began tracking this model

Quality & price over time

Since 25 Jul 2026

Known Good Index 47 · based on 4 of 7 evaluations
Blended price $0.17 per 1M · observed 9 Sep 2026
Output speed 21 tok/s · observed 9 Sep 2026
Time to first token 1.6s

No recorded changes for this model since tracking began.

The Known Good

Comparisons

Head-to-head pages pairing Google: Gemma 3 27B with other tracked models

Google: Gemma 3 27B vs OpenAI: GPT-4.1 Google: Gemma 3 27B vs Anthropic: Claude Opus 4.1 Google: Gemma 3 27B vs DeepSeek-R1-Distill-Qwen-32B All 49 comparisons →
Sources for this page
Pricing & providers — OpenRouter Metadata — models.dev Benchmarks — Epoch AI (CC-BY 4.0)Benchmarks — LiveBench (Apache-2.0)Elo — Arena Full attribution →