Google model product
Gemma 4 26B A4B
8 billion activated parameters per token.
Updated Sep 8, 2026. Default version: Gemma 4 26B-A4B
LLMBoard Score51.0Gemma 4 26B-A4B
Coverage40%12 benchmark families
Context window262.1KTokens
Official input priceN/AOfficial price unavailable
Gemma 4 26B A4B Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Gemma 4 26B A4B Benchmark Results
Benchmark scores for Gemma 4 26B-A4B.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkBIG-Bench Extra Hard | Score64.80% | Rank02 | Participants11 | Percentile90.00% | EvidenceC | Evaluated |
| BenchmarkMedXpertQA | Score58.10% | Rank06 | Participants12 | Percentile54.55% | EvidenceC | Evaluated |
| Benchmarkt2-bench | Score85.50% | Rank06 | Participants23 | Percentile77.27% | EvidenceC | Evaluated |
| BenchmarkMRCR v2 (8-needle) | Score44.10% | Rank11 | Participants24 | Percentile56.52% | EvidenceC | Evaluated |
| BenchmarkMathVision | Score82.40% | Rank15 | Participants35 | Percentile58.82% | EvidenceC | Evaluated |
| BenchmarkAIME 2026 | Score88.30% | Rank17 | Participants22 | Percentile23.81% | EvidenceC | Evaluated |
| BenchmarkMMMLU | Score86.30% | Rank24 | Participants49 | Percentile52.08% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench v6 | Score77.10% | Rank28 | Participants62 | Percentile55.74% | EvidenceC | Evaluated |
| BenchmarkLM Arena Vision | Score1,259.45 rating | Rank36 | Participants107 | Percentile66.98% | EvidenceA | Evaluated |
| BenchmarkMMMU-Pro | Score73.80% | Rank37 | Participants69 | Percentile47.06% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score82.60% | Rank38 | Participants138 | Percentile72.99% | EvidenceC | Evaluated |
| BenchmarkLM Arena Vision Style Control | Score1,241.54 rating | Rank39 | Participants107 | Percentile64.15% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Factuality | Score1,445.10 rating | Rank52 | Participants121 | Percentile57.50% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text | Score1,434.86 rating | Rank63 | Participants210 | Percentile70.33% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,438.45 rating | Rank67 | Participants210 | Percentile68.42% | EvidenceA | Evaluated |
| BenchmarkLM Arena Webdev | Score1,361.20 rating | Rank70 | Participants97 | Percentile28.13% | EvidenceA | Evaluated |
| BenchmarkHumanity's Last Exam | Score17.20% | Rank74 | Participants103 | Percentile28.43% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score82.30% | Rank82 | Participants247 | Percentile67.07% | EvidenceC | Evaluated |
Gemma 4 26B A4B Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenavision | Categorycaptioning | Rank04 | Rating / score1,284.24 | Votes49 | ObservationsN/A | Result date |
| Arenavision style control | Categorycaptioning | Rank07 | Rating / score1,227.93 | Votes49 | ObservationsN/A | Result date |
| Arenavision | Categoryhomework | Rank23 | Rating / score1,293.44 | Votes3,671 | ObservationsN/A | Result date |
| Arenavision | Categoryhumor | Rank26 | Rating / score1,271.15 | Votes931 | ObservationsN/A | Result date |
| Arenavision style control | Categoryhomework | Rank26 | Rating / score1,285.57 | Votes3,671 | ObservationsN/A | Result date |
| Arenatext | Categorymath | Rank27 | Rating / score1,467.90 | Votes374 | ObservationsN/A | Result date |
| Arenavision | Categorychinese | Rank28 | Rating / score1,315.67 | Votes1,405 | ObservationsN/A | Result date |
| Arenavision style control | Categorychinese | Rank29 | Rating / score1,295.14 | Votes1,405 | ObservationsN/A | Result date |
| Arenavision style control | Categoryhumor | Rank31 | Rating / score1,244.76 | Votes931 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank32 | Rating / score1,470.64 | Votes299 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry mathematical | Rank32 | Rating / score1,475.69 | Votes299 | ObservationsN/A | Result date |
| Arenavision | Categorycreative writing vision | Rank34 | Rating / score1,253.56 | Votes1,399 | ObservationsN/A | Result date |
| Arenavision | Categoryocr | Rank34 | Rating / score1,267.73 | Votes16,009 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank35 | Rating / score1,466.99 | Votes251 | ObservationsN/A | Result date |
| Arenatext style control | Categorymath | Rank35 | Rating / score1,468.28 | Votes374 | ObservationsN/A | Result date |
| Arenavision | Categoryoverall | Rank36 | Rating / score1,259.45 | Votes22,710 | ObservationsN/A | Result date |
| Arenavision style control | Categoryentity recognition | Rank36 | Rating / score1,148.39 | Votes206 | ObservationsN/A | Result date |
| Arenavision | Categoryentity recognition | Rank37 | Rating / score1,179.20 | Votes206 | ObservationsN/A | Result date |
| Arenavision style control | Categoryocr | Rank39 | Rating / score1,257.32 | Votes16,009 | ObservationsN/A | Result date |
| Arenavision style control | Categoryoverall | Rank39 | Rating / score1,241.54 | Votes22,710 | ObservationsN/A | Result date |
| Arenavision | Categorydiagram | Rank40 | Rating / score1,263.62 | Votes5,907 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry legal and government | Rank41 | Rating / score1,456.70 | Votes391 | ObservationsN/A | Result date |
| Arenatext style control | Categoryfrench | Rank41 | Rating / score1,478.21 | Votes251 | ObservationsN/A | Result date |
| Arenavision style control | Categorycreative writing vision | Rank41 | Rating / score1,227.53 | Votes1,399 | ObservationsN/A | Result date |
| Arenavision | Categoryenglish | Rank42 | Rating / score1,254.90 | Votes9,228 | ObservationsN/A | Result date |
| Arenavision style control | Categoryenglish | Rank44 | Rating / score1,237.45 | Votes9,228 | ObservationsN/A | Result date |
| Arenavision style control | Categorydiagram | Rank45 | Rating / score1,260.89 | Votes5,907 | ObservationsN/A | Result date |
| Arenatext | Categoryrussian | Rank46 | Rating / score1,437.57 | Votes820 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry legal and government | Rank46 | Rating / score1,463.44 | Votes391 | ObservationsN/A | Result date |
| Arenatext | Categorychinese | Rank47 | Rating / score1,482.76 | Votes468 | ObservationsN/A | Result date |
Gemma 4 26B A4B Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- From $0.042 input, $0.22 output per 1M via Kilo Gateway
- Tracked offerings
- 18
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderKilo Gateway | Provider model IDgoogle/gemma-4-26b-a4b-it | Regionglobal | Input / 1M$0.042 | Output / 1M$0.22 | Context262.1K | Updated |
| ProviderEmpirioLabs AI | Provider model IDgemma-4-26b-a4b | Regionglobal | Input / 1M$0.05 | Output / 1M$0.29 | Context262.1K | Updated |
| ProviderOrcaRouter | Provider model IDgoogle/gemma-4-26b-a4b-it | Regionglobal | Input / 1M$0.06 | Output / 1M$0.33 | Context262.1K | Updated |
| ProviderDeep Infra | Provider model IDgoogle/gemma-4-26B-A4B-it | Regionglobal | Input / 1M$0.07 | Output / 1M$0.34 | Context262.1K | Updated |
| ProviderOpenRouter | Provider model IDgoogle/gemma-4-26b-a4b-it | Regionglobal | Input / 1M$0.07 | Output / 1M$0.34 | Context262.1K | Updated |
| ProviderRequesty | Provider model IDgemma-4-26b-a4b-it | Regionglobal | Input / 1M$0.07 | Output / 1M$0.34 | Context262.1K | Updated |
| ProviderDevPass (LLM Gateway) | Provider model IDgemma-4-26b-a4b-it | Regionglobal | Input / 1M$0.07 | Output / 1M$0.34 | Context262.1K | Updated |
| ProviderCharm Hyper | Provider model IDgemma-4-26b-a4b-it | Regionglobal | Input / 1M$0.11 | Output / 1M$0.408 | Context256K | Updated |
| ProviderCortecs | Provider model IDgemma-4-26b-a4b-it | Regionglobal | Input / 1M$0.111 | Output / 1M$0.557 | Context262K | Updated |
| ProviderNanoGPT | Provider model IDgoogle/gemma-4-26b-a4b-it | Regionglobal | Input / 1M$0.12 | Output / 1M$0.38 | Context262.1K | Updated |
| ProviderSiliconFlow | Provider model IDgoogle/gemma-4-26B-A4B-it | Regionglobal | Input / 1M$0.12 | Output / 1M$0.40 | Context262.1K | Updated |
| ProviderHugging Face | Provider model IDgoogle/gemma-4-26B-A4B-it | Regionglobal | Input / 1M$0.13 | Output / 1M$0.40 | Context262.1K | Updated |
| ProviderMerge Gateway | Provider model IDgoogle/gemma-4-26b-a4b-it | Regionglobal | Input / 1M$0.13 | Output / 1M$0.40 | Context262.1K | Updated |
| ProviderNovitaAI | Provider model IDgoogle/gemma-4-26b-a4b-it | Regionglobal | Input / 1M$0.13 | Output / 1M$0.40 | Context262.1K | Updated |
| Providerevroc | Provider model IDgoogle/gemma-4-26B-A4B-it | Regionglobal | Input / 1M$0.144 | Output / 1M$0.575 | Context262.1K | Updated |
| ProviderVercel AI Gateway | Provider model IDgoogle/gemma-4-26b-a4b-it | Regionglobal | Input / 1M$0.15 | Output / 1M$0.60 | Context262.1K | Updated |
| ProviderScaleway | Provider model IDgemma-4-26b-a4b-it | Regionglobal | Input / 1M$0.25 | Output / 1M$0.50 | Context256K | Updated |
Gemma 4 26B A4B Runtime Performance
Provider-specific output speed and catalog latency for Gemma 4 26B-A4B. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Gemma 4 26B A4B Specifications
Technical details for the model's default version.
- Version
- Gemma 4 26B-A4B
- Released
- Apr 2, 2026
- Knowledge cutoff
- Jan 1, 2025
- Parameters
- 25.2B
- Context window
- 262.1K
- Max output
- 131.1K
- Inputs
- image, text
- Outputs
- text
- Open weights
- Yes
- License
- Apache 2.0
Gemma 4 26B A4B Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGemma 4 26B-A4B | Released | LLMBoard51.03 | Parameters25.2B | Context262.1K | Max output131.1K | Open weightsYes | LicenseApache 2.0 |
Models similar to Gemma 4 26B A4B
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
GO
Google
Gemini 3.1 Flash Lite
GOGoogle
Gemini 2.5 Pro
GOGoogle
Gemini 3.5 Flash Lite
GOGoogle
Gemma 4 31B
GOGoogle
Gemini 2.5 Flash
GOGoogle
Gemini 3.5 Flash Cyber
What is Gemma 4 26B A4B?
Key information about Gemma 4 26B A4B and its available data.
Gemma 4 26B A4B is Google DeepMind's multimodal Mixture-of-Experts model, supporting image and text inputs with a 256K context window. 8 billion activated parameters per token.
Data as of 2026-09-08.
FAQ
Common questions about Gemma 4 26B A4B.
When was Gemma 4 26B A4B released?
Gemma 4 26B A4B's default version was released on Apr 2, 2026.
How much does Gemma 4 26B A4B cost?
No official standard PAYG price is currently available for Gemma 4 26B A4B. The lowest tracked third-party offer starts at $0.042 input and $0.22 output via Kilo Gateway.
Who created Gemma 4 26B A4B?
Gemma 4 26B A4B was created by Google.
What is the context window for Gemma 4 26B A4B?
The default version has a 262.1K token context window.
Is Gemma 4 26B A4B open weight?
Yes. The default version is marked as open weight under Apache 2.0.
How many API providers offer Gemma 4 26B A4B?
18 provider offerings are linked to the default version.