The Known Good Updated 25 Jul 2026
Google: Gemma 3 27B vs OpenAI: gpt-oss-120b ·Google vs OpenAI

Google: Gemma 3 27B vs OpenAI: gpt-oss-120b — head-to-head comparison

Google: Gemma 3 27B profile OpenAI: gpt-oss-120b profile All comparisons
Google: Gemma 3 27B — Google
Quality
#47 of 61
37based on 3 of 6 evaluations
Known Good Index
Speed
#262 of 316
24.4
Output tokens per second
Price
#64 of 340
$0.17
Blended per 1M tokens
In $0.08 · Out $0.45
Cache hit price
#62 of 206
$0.040
per 1M tokens (−50%)
Context
#171 of 345
262k
tokens (~400 pages)
OpenAI: gpt-oss-120b — OpenAI
Quality
#33 of 61
61based on 3 of 6 evaluations
Known Good Index
Speed
#10 of 316
172.2
Output tokens per second
Price
#33 of 340
$0.07
Blended per 1M tokens
In $0.04 · Out $0.17
Cache hit price
#25 of 206
$0.015
per 1M tokens (−59%)
Context
#238 of 345
131k
tokens (~200 pages)
Summary

On the Known Good Index, Google: Gemma 3 27B scores 37 and OpenAI: gpt-oss-120b scores 61 — a 25-point lead for OpenAI: gpt-oss-120b. They share 2 published evaluations, with Google: Gemma 3 27B ahead on 0 and OpenAI: gpt-oss-120b ahead on 2.

Blended pricing is $0.17 per 1M tokens for Google: Gemma 3 27B and $0.07 for OpenAI: gpt-oss-120b — OpenAI: gpt-oss-120b is about 2.5× cheaper.

Throughput is 24 tokens per second for Google: Gemma 3 27B and 172 for OpenAI: gpt-oss-120b, with time to first token at 0.9s and 0.7s. Google: Gemma 3 27B accepts 262k tokens of context against 131k.

Google: Gemma 3 27B is open weights under Open weights (restricted use); OpenAI: gpt-oss-120b is open weights under Open weights (unrestricted). 17 tracked hosts serve one or both, and 5 serve both.

Generated from tracked data on 25 Jul 2026 · how these comparisons are computed →

At a glance
MetricGoogle: Gemma 3 27BOpenAI: gpt-oss-120b
Known Good Index 37 based on 3 of 6 evaluations 61 based on 3 of 6 evaluations
Blended $/1M $0.17 $0.07
Output tok/s 24 172
Context window 262k 131k
Providers 5 17
Jump to section

Quality

Known Good Index and the capability indices, side by side

IndicesComponents
Index scores

Each index is min-max normalised across tracked models and scaled 0–100 from published evaluations. We do not run these evaluations.

MetricGoogle: Gemma 3 27BOpenAI: gpt-oss-120b
Known Good Index 37 61
Mathematics Index 47 89
Science Index 44 77
Reasoning Index 40
Agentic Index 18
Openness Index 60 60
The Known Good

Quality Breakdown

Every evaluation both models report a published score for

Evaluations both models report

Bar = Google: Gemma 3 27B · marker = OpenAI: gpt-oss-120b · scores ingested from Epoch AI (CC-BY 4.0)

GPQA Diamond49 / 76
MATH / AIME20 / 89

Right-hand figures read Google: Gemma 3 27B / OpenAI: gpt-oss-120b. Google: Gemma 3 27B leads on 0 of 2, OpenAI: gpt-oss-120b on 2.

The Known Good

Arena Elo

Human preference rating from blind pairwise votes

Arena ratings

Elo from blind pairwise votes, ingested from Arena. Higher is better.

MetricGoogle: Gemma 3 27BOpenAI: gpt-oss-120b
Text Arena 1,366
Text Arena (style controlled) 1,352
The Known Good

Price & Cost

Published API pricing by token type

Per token typeCache discount
Published pricing

USD per 1M tokens. Blended is a 3:1 input:output mix. Live from OpenRouter, cross-checked against models.dev. Lower is better.

MetricGoogle: Gemma 3 27BOpenAI: gpt-oss-120b
Input $/1M $0.08 $0.04
Output $/1M $0.45 $0.17
Blended $/1M $0.17 $0.07
Cache read $/1M $0.040 $0.015
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 accepted and maximum output emitted

Context and output limits

Provider-declared limits, from metadata via models.dev and OpenRouter. Higher is better.

MetricGoogle: Gemma 3 27BOpenAI: gpt-oss-120b
Context window 262k 131k
Max output tokens 131k 131k
The Known Good

Speed & Latency

Median throughput and time to first token across hosting providers

Output speedTime to first token
Throughput and latency

Medians across all tracked providers, derived from OpenRouter endpoint telemetry.

MetricGoogle: Gemma 3 27BOpenAI: gpt-oss-120b
Output tokens/s 24 172
Time to first token 0.9s 0.7s
The Known Good

Providers

Which hosts serve each model, and at what price and speed

17 hosts · 5 serve both
Full provider benchmarks →
Provider Google: Gemma 3 27B blendedGoogle: Gemma 3 27B tok/s OpenAI: gpt-oss-120b blendedOpenAI: gpt-oss-120b tok/s
Parasail $0.17 54 $0.26 207
Novita $0.14 32 $0.10 42
DeepInfra $0.10 22 $0.07 246
Phala $0.23 15 $0.26 82
Nebius $0.15 12 $0.26 281
Cerebras $0.45 495
Groq $0.26 411
SambaNova $0.34 306
BaseTen $0.20 285
Mara $0.30 181
Amazon Bedrock · first party $0.26 166
Together $0.26 111
Google · first party $0.16 85
DigitalOcean $0.17 62
CoreWeave $0.07 41
Mancer 2 $0.17 41
SiliconFlow $0.15 26

Per-host pricing and telemetry from OpenRouter. A dash means that host does not serve that model.

The Known Good
Sources for this page
Pricing & providers — OpenRouter Metadata — models.dev Benchmarks — Epoch AI (CC-BY 4.0)Elo — Arena Full attribution →