Meta model product
Llama 4 Maverick
Llama 4 Maverick is a natively multimodal model capable of processing both text and images.
Updated Aug 17, 2026. Default version: Llama 4 Maverick
LLMBoard Score24.2Llama 4 Maverick
Coverage80%12 benchmark families
Context window1MTokens
Official input priceN/AOfficial price unavailable
Llama 4 Maverick Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Llama 4 Maverick Benchmark Results
Benchmark scores for Llama 4 Maverick.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkMGSM | Score92.3% | Rank01 | Participants31 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkTydiQA | Score31.7% | Rank01 | Participants2 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkChartQA | Score90.0% | Rank02 | Participants26 | Percentile96.0% | EvidenceC | Evaluated |
| BenchmarkDocVQA | Score94.4% | Rank07 | Participants28 | Percentile77.8% | EvidenceC | Evaluated |
| BenchmarkMathVista | Score73.7% | Rank08 | Participants39 | Percentile81.6% | EvidenceC | Evaluated |
| BenchmarkMBPP | Score77.6% | Rank16 | Participants33 | Percentile53.1% | EvidenceC | Evaluated |
| BenchmarkMMMU | Score73.4% | Rank25 | Participants63 | Percentile61.3% | EvidenceC | Evaluated |
| BenchmarkMMLU | Score85.5% | Rank39 | Participants101 | Percentile62.0% | EvidenceC | Evaluated |
| BenchmarkMATH | Score61.2% | Rank46 | Participants71 | Percentile35.7% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench | Score43.4% | Rank51 | Participants75 | Percentile32.4% | EvidenceC | Evaluated |
| BenchmarkMMMU-Pro | Score59.6% | Rank53 | Participants68 | Percentile22.4% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score80.5% | Rank57 | Participants134 | Percentile57.9% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score69.8% | Rank136 | Participants239 | Percentile43.3% | EvidenceC | Evaluated |
Llama 4 Maverick Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenavision | Categoryoverall | Rank99 | Rating / score1143.0 | Votes6,998 | ObservationsN/A | Result date |
| Arenavision style control | Categoryoverall | Rank99 | Rating / score1146.8 | Votes6,998 | ObservationsN/A | Result date |
| Arenatext style control | Categoryoverall | Rank226 | Rating / score1327.0 | Votes40,065 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank234 | Rating / score1287.5 | Votes40,065 | ObservationsN/A | Result date |
Llama 4 Maverick Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- From $0.15 input, $0.60 output per 1M via NanoGPT
- Tracked offerings
- 5
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderNanoGPT | Provider model IDmeta-llama/llama-4-maverick | Regionglobal | Input / 1M$0.15 | Output / 1M$0.60 | Context1M | Updated |
| ProviderHelicone | Provider model IDllama-4-maverick | Regionglobal | Input / 1M$0.15 | Output / 1M$0.60 | Context131.1K | Updated |
| ProviderOpenRouter | Provider model IDmeta-llama/llama-4-maverick | Regionglobal | Input / 1M$0.20 | Output / 1M$0.80 | Context1M | Updated |
| ProviderDigitalOcean | Provider model IDllama-4-maverick | Regionglobal | Input / 1M$0.20 | Output / 1M$0.696 | Context128K | Updated |
| ProviderNeon | Provider model IDllama-4-maverick | Regionglobal | Input / 1M$0.50 | Output / 1M$1.5 | Context1M | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Llama 4 Maverick Runtime Performance
Provider-specific output speed and catalog latency for Llama 4 Maverick. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderSambanova | Output Speed638.7 tok/s | Catalog Latency2.04 s | Max Input1M | Max Output1M | Updated |
| ProviderGroq | Output Speed307.3 tok/s | Catalog Latency0.27 s | Max Input1M | Max Output1M | Updated |
| ProviderTogether | Output Speed97.93 tok/s | Catalog Latency0.2 s | Max Input1M | Max Output1M | Updated |
| ProviderLambda | Output Speed93.69 tok/s | Catalog Latency0.65 s | Max Input1M | Max Output1M | Updated |
| ProviderDeepInfra | Output Speed83.59 tok/s | Catalog Latency0.38 s | Max Input1M | Max Output1M | Updated |
| ProviderNovita | Output Speed69.42 tok/s | Catalog Latency0.62 s | Max Input1M | Max Output1M | Updated |
| ProviderFireworks | Output Speed63.03 tok/s | Catalog Latency0.62 s | Max Input1M | Max Output1M | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Llama 4 Maverick Specifications
Technical details for the model's default version.
- Version
- Llama 4 Maverick
- Released
- Apr 5, 2025
- Knowledge cutoff
- Unknown
- Parameters
- 400B
- Context window
- 1M
- Max output
- 1M
- Inputs
- image, text
- Outputs
- text
- Open weights
- No
- License
- Llama 4 Community License Agreement
Llama 4 Maverick 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 |
|---|
| VersionLlama 4 Maverick | Released | LLMBoard24.2 | Parameters400B | Context1M | Max output1M | Open weightsNo | LicenseLlama 4 Community License Agreement |
Llama 4 Maverick vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Llama 4 Maverick
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
ME
Llama 3.1 405B
Meta
ME
Llama 3.3 70B
Meta
ME
Llama 4 Scout
Meta
ME
Llama 3.1 70B
Meta
ME
Llama 3.2 90B
Meta
ME
Llama 3.2 3B
Meta
What is Llama 4 Maverick?
Key information about Llama 4 Maverick and its available data.
Llama 4 Maverick is a natively multimodal model capable of processing both text and images. It features a 17 billion active parameter mixture-of-experts (MoE) architecture with 128 experts, supporting a wide range of multimodal tasks such as conversational interaction, image analysis, and code generation.
The model includes a 1 million token context window.
Data as of 2026-08-17.
FAQ
Common questions about Llama 4 Maverick.
When was Llama 4 Maverick released?
Llama 4 Maverick's default version was released on Apr 5, 2025.
How much does Llama 4 Maverick cost?
No official standard PAYG price is currently available for Llama 4 Maverick. The lowest tracked third-party offer starts at $0.15 input and $0.60 output via NanoGPT.
Who created Llama 4 Maverick?
Llama 4 Maverick was created by Meta.
What is the context window for Llama 4 Maverick?
The default version has a 1M token context window.
Is Llama 4 Maverick open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer Llama 4 Maverick?
5 provider offerings are linked to the default version.
What models should I compare Llama 4 Maverick with?
Nearby ranked alternatives include Phi 4 Reasoning Plus, DeepSeek-V3, Qwen2 VL 72B.