Meta model product
Llama 4 Maverick
Llama 4 Maverick is a natively multimodal model from Meta that processes text and images.
Updated Oct 23, 2026. Default version: Llama 4 Maverick
LLMBoard Score22.3Llama 4 Maverick
Coverage100%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 | Score92.30% | Rank01 | Participants31 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkTydiQA | Score31.70% | Rank01 | Participants2 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkChartQA | Score90.00% | Rank02 | Participants26 | Percentile96.00% | EvidenceC | Evaluated |
| BenchmarkDocVQA | Score94.40% | Rank07 | Participants28 | Percentile77.78% | EvidenceC | Evaluated |
| BenchmarkMathVista | Score73.70% | Rank09 | Participants40 | Percentile79.49% | EvidenceC | Evaluated |
| BenchmarkMBPP | Score77.60% | Rank16 | Participants33 | Percentile53.13% | EvidenceC | Evaluated |
| BenchmarkMMMU | Score73.40% | Rank28 | Participants66 | Percentile58.46% | EvidenceC | Evaluated |
| BenchmarkMMLU | Score85.50% | Rank40 | Participants103 | Percentile61.76% | EvidenceC | Evaluated |
| BenchmarkMATH | Score61.20% | Rank46 | Participants71 | Percentile35.71% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench | Score43.40% | Rank52 | Participants76 | Percentile32.00% | EvidenceC | Evaluated |
| BenchmarkMMMU-Pro | Score59.60% | Rank58 | Participants73 | Percentile20.83% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score80.50% | Rank61 | Participants142 | Percentile57.45% | EvidenceC | Evaluated |
| BenchmarkAA SciCode Subtasks | Score31.71% | Rank77 | Participants89 | Percentile13.64% | EvidenceB | EvaluatedN/A |
| BenchmarkAA AIME 2025 | Score19.33% | Rank80 | Participants96 | Percentile16.84% | EvidenceB | EvaluatedN/A |
| BenchmarkAA Omniscience Accuracy | Score24.90% | Rank88 | Participants201 | Percentile56.50% | EvidenceB | EvaluatedN/A |
| BenchmarkLM Arena Vision | Score1,141.71 rating | Rank91 | Participants113 | Percentile19.64% | EvidenceA | Evaluated |
| BenchmarkLM Arena Vision Style Control | Score1,145.45 rating | Rank92 | Participants113 | Percentile18.75% | EvidenceA | Evaluated |
| BenchmarkAA IFBench | Score42.99% | Rank102 | Participants168 | Percentile39.52% | EvidenceB | EvaluatedN/A |
| BenchmarkAA LCR v1.1 | Score50.00% | Rank127 | Participants193 | Percentile34.38% | EvidenceB | EvaluatedN/A |
| BenchmarkAA GPQA Diamond | Score67.07% | Rank141 | Participants199 | Percentile29.29% | EvidenceB | EvaluatedN/A |
| BenchmarkGPQA | Score69.80% | Rank150 | Participants256 | Percentile41.57% | EvidenceC | Evaluated |
| BenchmarkAA HLE Text No Tools | Score4.91% | Rank160 | Participants200 | Percentile20.10% | EvidenceB | EvaluatedN/A |
| BenchmarkLM Arena Text Style Control | Score1,327.01 rating | Rank172 | Participants218 | Percentile21.20% | EvidenceA | Evaluated |
| BenchmarkAA CritPt | Score0.00% | Rank173 | Participants200 | Percentile13.57% | EvidenceB | EvaluatedN/A |
| BenchmarkLM Arena Text | Score1,287.24 rating | Rank176 | Participants218 | Percentile19.35% | EvidenceA | 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 style control | Categorycreative writing | Rank29 | Rating / score1,127.80 | Votes250 | ObservationsN/A | Result date |
| Arenavision | Categorycreative writing | Rank30 | Rating / score1,112.91 | Votes250 | ObservationsN/A | Result date |
| Arenavision style control | Categoryhumor | Rank74 | Rating / score1,097.71 | Votes216 | ObservationsN/A | Result date |
| Arenavision | Categoryhumor | Rank75 | Rating / score1,100.04 | Votes216 | ObservationsN/A | Result date |
| Arenavision style control | Categorycreative writing vision | Rank80 | Rating / score1,110.98 | Votes214 | ObservationsN/A | Result date |
| Arenavision | Categorychinese | Rank81 | Rating / score1,140.83 | Votes294 | ObservationsN/A | Result date |
| Arenavision | Categorycreative writing vision | Rank81 | Rating / score1,099.04 | Votes214 | ObservationsN/A | Result date |
| Arenavision | Categoryhomework | Rank82 | Rating / score1,164.18 | Votes526 | ObservationsN/A | Result date |
| Arenavision style control | Categorychinese | Rank83 | Rating / score1,147.33 | Votes294 | ObservationsN/A | Result date |
| Arenavision style control | Categoryhomework | Rank85 | Rating / score1,164.70 | Votes526 | ObservationsN/A | Result date |
| Arenavision | Categoryocr | Rank87 | Rating / score1,142.94 | Votes3,083 | ObservationsN/A | Result date |
| Arenavision | Categorydiagram | Rank88 | Rating / score1,133.52 | Votes947 | ObservationsN/A | Result date |
| Arenavision style control | Categorydiagram | Rank88 | Rating / score1,152.33 | Votes947 | ObservationsN/A | Result date |
| Arenavision style control | Categoryocr | Rank88 | Rating / score1,153.13 | Votes3,083 | ObservationsN/A | Result date |
| Arenavision | Categoryoverall | Rank91 | Rating / score1,141.71 | Votes6,923 | ObservationsN/A | Result date |
| Arenavision style control | Categoryoverall | Rank92 | Rating / score1,145.45 | Votes6,923 | ObservationsN/A | Result date |
| Arenavision | Categoryenglish | Rank93 | Rating / score1,149.99 | Votes3,387 | ObservationsN/A | Result date |
| Arenavision style control | Categoryenglish | Rank93 | Rating / score1,150.87 | Votes3,387 | ObservationsN/A | Result date |
| Arenatext style control | Categorygerman | Rank139 | Rating / score1,331.32 | Votes1,028 | ObservationsN/A | Result date |
| Arenatext style control | Categorypolish | Rank139 | Rating / score1,330.22 | Votes2,675 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank145 | Rating / score1,257.32 | Votes935 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank147 | Rating / score1,291.17 | Votes1,028 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank149 | Rating / score1,275.23 | Votes2,675 | ObservationsN/A | Result date |
| Arenatext style control | Categorykorean | Rank153 | Rating / score1,262.06 | Votes796 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank154 | Rating / score1,206.31 | Votes935 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank154 | Rating / score1,203.29 | Votes796 | ObservationsN/A | Result date |
| Arenatext | Categoryspanish | Rank154 | Rating / score1,293.12 | Votes820 | ObservationsN/A | Result date |
| Arenatext style control | Categoryspanish | Rank157 | Rating / score1,325.94 | Votes820 | ObservationsN/A | Result date |
| Arenatext style control | Categorycreative writing | Rank159 | Rating / score1,306.93 | Votes5,152 | ObservationsN/A | Result date |
| Arenatext style control | Categoryrussian | Rank161 | Rating / score1,327.74 | Votes3,104 | 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 Helicone
- Tracked offerings
- 5
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderHelicone | Provider model IDllama-4-maverick | Regionglobal | Input / 1M$0.15 | Output / 1M$0.60 | Context131.1K | Updated |
| ProviderNanoGPT | Provider model IDmeta-llama/llama-4-maverick | Regionglobal | Input / 1M$0.1875 | Output / 1M$0.6525 | Context1M | Updated |
| ProviderOpenRouter | Provider model IDmeta-llama/llama-4-maverick | Regionglobal | Input / 1M$0.1875 | Output / 1M$0.6525 | Context524.3K | Updated |
| ProviderDigitalOcean | Provider model IDllama-4-maverick | Regionglobal | Input / 1M$0.25 | Output / 1M$0.87 | 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 Speed638.70 tok/s | Catalog Latency2.04 s | Max Input1M | Max Output1M | Updated |
| ProviderGroq | Output Speed307.30 tok/s | Catalog Latency0.27 s | Max Input1M | Max Output1M | Updated |
| ProviderDeepInfra | Output Speed83.59 tok/s | Catalog Latency0.38 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
- Yes
- 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 | LLMBoard22.27 | Parameters400B | Context1M | Max output1M | Open weightsYes | LicenseLlama 4 Community License Agreement |
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 3.1 8B
Meta
What is Llama 4 Maverick?
Key information about Llama 4 Maverick and its available data.
Llama 4 Maverick is a 17 billion active parameter mixture-of-experts (MoE) model with 128 experts, designed for conversational interaction, image analysis, and code generation. It has a 1 million token context window.
Data as of 2026-10-22.
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 Helicone.
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?
Yes. The default version is marked as open weight under Llama 4 Community License Agreement.