LLM Pricing
PricingLeaderboardsToolsProvidersReleasesGuides

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list

Gemma 3 27B IT

google·google/gemma-3-27b-it·GA·Open weights·gemma series
Largest open Gemma 3 instruction model for multilingual text generation and visual understanding

Gemma 3 27B IT by google is offered by 16 providers on this page. Public prices are shown for 15 of them. Its reference price is $0.53 per 1M input tokens and $0.76 per 1M output tokens. The lowest paid channel is Deep Infra at $0.08 / $0.16 per 1M, about 6.6× below the reference price. It also has 1 free ($0) channel; free tiers usually carry rate limits, and subscription-covered access bills $0 per token only after the subscription fee.

Artificial Analysis rates it 4.9 on the Intelligence Index, with 10.1 for coding and 0.1 for agentic tasks. Against its lowest blended price of $0.10 per 1M, that is roughly 49 index points per dollar, which is the value ratio the leaderboards rank on.

The context window is 37,000 tokens at the reference host, but hosts report different limits, from 37,000 to 131,072, so the usable window depends on the provider you pick. It supports tool use and structured output. Accepted input modalities are Text, Image, and PDF. The weights are open, so it can also be self-hosted or served through a gateway of your choice. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use. Its training knowledge cuts off at 2024-08.

Specs & pricing

Input / output per 1M tokens
Reference price·STACKIT
$0.53 / $0.76
Blended $0.59 · Cache read —
Lowest paid·Deep InfraCloud
$0.08 / $0.16
Blended $0.10 · 6.6× spread
1 more $0 channels
Context
37,000
Output limit
4,096
Knowledge cutoff
2024-08
Released / updated
2025-03-12 / 2025-03-12
Capabilities
Reasoning✓ Tool use✓ Structured output✓ Temperature✓ Attachments
⚠ Providers report capability flags inconsistently
Modalities
TextImagePDF
Weights
Hugging Face

Quality & performanceArtificial Analysis · Intelligence Index v4.3

Intelligence4.9
Value49
Coding10.1
Agentic0.1
Output speed—
TTFT—
Value formulaIQ 4.9 ÷ min blended $0.100 = 49

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 16 providers15 with public prices · 1 free

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
EmpirioLabs AI
gemma-3-27b
GatewayFree——128,000 ⚠8,192
Deep InfraCloud$0.08$0.16——131,072 ⚠131,072
Kilo GatewayGateway$0.08$0.16——131,072 ⚠117,964
Hugging FaceCloud$0.08$0.16——131,072 ⚠131,072
Eden AIGateway$0.08$0.16——131,072 ⚠131,072
NovitaAIGateway$0.12$0.20——98,304 ⚠16,384
CortecsGateway$0.099$0.30——131,000 ⚠110,000
Nebius Token FactoryCloud$0.10$0.30$0.01$0.13110,000 ⚠8,192
Eden AIGateway$0.10$0.30$0.10—110,000 ⚠—
OpenRouterGateway$0.08$0.45$0.04—131,072 ⚠117,964

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Your usage cost

1Deep Infra$24.00
2Kilo Gateway$24.00
3Hugging Face$24.00
4Eden AI$24.00
5Nebius Token Factory$24.20
6OpenRouter$33.70

1 more channels offer $0 (EmpirioLabs AI); free tiers usually have rate limits and no SLA, excluded from ranking.

The cheapest paid channel is the only channel.

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$0.80$02026-08-052026-09-032026-08-05 · Input list $0.532026-08-06 · Input list $0.532026-08-07 · Input list $0.532026-08-08 · Input list $0.532026-08-09 · Input list $0.532026-08-10 · Input list $0.532026-08-11 · Input list $0.532026-08-12 · Input list $0.532026-08-13 · Input list $0.202026-08-29 · Input list $0.532026-09-03 · Input list $0.532026-08-05 · Output list $0.762026-08-06 · Output list $0.762026-08-07 · Output list $0.762026-08-08 · Output list $0.762026-08-09 · Output list $0.762026-08-10 · Output list $0.762026-08-11 · Output list $0.762026-08-12 · Output list $0.762026-08-13 · Output list $0.802026-08-29 · Output list $0.762026-09-03 · Output list $0.762026-08-05 · Min blended $0.142026-08-06 · Min blended $0.142026-08-07 · Min blended $0.142026-08-08 · Min blended $0.142026-08-09 · Min blended $0.142026-08-10 · Min blended $0.142026-08-11 · Min blended $0.142026-08-12 · Min blended $0.142026-08-13 · Min blended $0.352026-08-29 · Min blended $0.142026-09-03 · Min blended $0.10

Artificial Analysis evaluations14 items

GPQA Diamond42.8%
Humanity's Last Exam4.4%
MMLU-Pro66.9%
SciCode23.3%
LiveCodeBench13.7%
Terminal-Bench Hard3.8%
Terminal-Bench 2.14.5%
AIME 202520.7%
AIME25.3%
MATH-50088.3%
τ²-Bench Telecom10.5%
τ³-Bench Banking0.8%
AA-LCR7.3%
IFBench31.8%

Individual evaluations run by Artificial Analysis, on the same reasoning tier as the intelligence score above. Each benchmark has its own task set and harness, so rows are not comparable with one another. The Intelligence Index above draws on a different, newer set of evaluations.

Related models

Gemma 4 26B A4B Cybersecuritysame series$0.11 / $0.33Gemma 4 31B Split-Untiedsame series$0.10 / $0.30Gemma 4 12B Semancersame series$0.05 / $0.25Command Rcheaper alternative$0.15 / $0.60Gemma 3 12B ITcheaper alternative$0.27 / $0.27Qwen3 30B A3B Instruct 2507cheaper alternative$0.30 / $0.50DeepSeek V3.2 Expcheaper alternative$0.29 / $0.43
Data partly from models.dev (MIT) · STACKIT official docs ↗