Meta model product
Llama 3.1 405B
1 405B Instruct is a large language model optimized for multilingual dialogue use cases.
Updated Sep 8, 2026. Default version: Llama 3.1 405B Instruct
LLMBoard Score22.9Llama 3.1 405B Instruct
Coverage40%17 benchmark families
Context window128KTokens
Official input priceN/AOfficial price unavailable
Llama 3.1 405B Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Llama 3.1 405B Benchmark Results
Benchmark scores for Llama 3.1 405B Instruct.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkAPI-Bank | Score92.00% | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkBFCL | Score88.50% | Rank01 | Participants11 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkGorilla Benchmark API Bench | Score35.30% | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkMBPP EvalPlus | Score88.60% | Rank01 | Participants2 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkMMLU (CoT) | Score88.60% | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkMultilingual MGSM (CoT) | Score91.60% | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkMultipl-E HumanEval | Score75.20% | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkMultipl-E MBPP | Score65.70% | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkNexus | Score58.70% | Rank01 | Participants4 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkARC-C | Score96.90% | Rank02 | Participants34 | Percentile96.97% | EvidenceC | Evaluated |
| BenchmarkGSM8k | Score96.80% | Rank05 | Participants48 | Percentile91.49% | EvidenceC | Evaluated |
| BenchmarkDROP | Score84.80% | Rank07 | Participants30 | Percentile79.31% | EvidenceC | Evaluated |
| BenchmarkHumanEval | Score89.00% | Rank17 | Participants66 | Percentile75.38% | EvidenceC | Evaluated |
| BenchmarkIFEval | Score88.60% | Rank26 | Participants68 | Percentile62.69% | EvidenceC | Evaluated |
| BenchmarkMMLU | Score87.30% | Rank26 | Participants101 | Percentile75.00% | EvidenceC | Evaluated |
| BenchmarkMATH | Score73.80% | Rank30 | Participants71 | Percentile58.57% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score73.30% | Rank83 | Participants138 | Percentile40.15% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,334.92 rating | Rank161 | Participants210 | Percentile23.45% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text | Score1,283.47 rating | Rank171 | Participants210 | Percentile18.66% | EvidenceA | Evaluated |
| BenchmarkGPQA | Score50.70% | Rank188 | Participants247 | Percentile23.98% | EvidenceC | Evaluated |
Llama 3.1 405B Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext style control | Categoryjapanese | Rank129 | Rating / score1,273.43 | Votes846 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank141 | Rating / score1,208.02 | Votes846 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry legal and government | Rank142 | Rating / score1,369.35 | Votes2,330 | ObservationsN/A | Result date |
| Arenatext style control | Categoryfrench | Rank145 | Rating / score1,340.31 | Votes364 | ObservationsN/A | Result date |
| Arenatext style control | Categorykorean | Rank145 | Rating / score1,254.16 | Votes924 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry life and physical and social science | Rank146 | Rating / score1,361.95 | Votes7,177 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry medicine and healthcare | Rank146 | Rating / score1,365.01 | Votes1,899 | ObservationsN/A | Result date |
| Arenatext style control | Categorygerman | Rank148 | Rating / score1,303.13 | Votes894 | ObservationsN/A | Result date |
| Arenatext style control | Categorymulti turn | Rank150 | Rating / score1,339.77 | Votes6,797 | ObservationsN/A | Result date |
| Arenatext style control | Categoryenglish | Rank151 | Rating / score1,361.19 | Votes23,318 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank152 | Rating / score1,278.39 | Votes364 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry entertainment and sports and media | Rank152 | Rating / score1,303.69 | Votes9,523 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry writing and literature and language | Rank152 | Rating / score1,314.94 | Votes15,999 | ObservationsN/A | Result date |
| Arenatext style control | Categorymath | Rank152 | Rating / score1,318.88 | Votes8,482 | ObservationsN/A | Result date |
| Arenatext style control | Categoryspanish | Rank152 | Rating / score1,311.18 | Votes348 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank153 | Rating / score1,183.66 | Votes924 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank154 | Rating / score1,252.09 | Votes894 | ObservationsN/A | Result date |
| Arenatext style control | Categorycreative writing | Rank154 | Rating / score1,304.02 | Votes8,948 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry mathematical | Rank154 | Rating / score1,319.53 | Votes7,242 | ObservationsN/A | Result date |
| Arenatext style control | Categoryhard prompts english | Rank155 | Rating / score1,361.03 | Votes6,791 | ObservationsN/A | Result date |
| Arenatext style control | Categorycoding | Rank156 | Rating / score1,375.99 | Votes6,249 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry software and it services | Rank157 | Rating / score1,363.39 | Votes10,298 | ObservationsN/A | Result date |
| Arenatext | Categoryspanish | Rank159 | Rating / score1,259.07 | Votes348 | ObservationsN/A | Result date |
| Arenatext style control | Categoryhard prompts | Rank159 | Rating / score1,341.31 | Votes10,652 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry legal and government | Rank160 | Rating / score1,303.35 | Votes3,383 | ObservationsN/A | Result date |
| Arenatext | Categorymulti turn | Rank160 | Rating / score1,297.03 | Votes6,797 | ObservationsN/A | Result date |
| Arenatext style control | Categoryexpert | Rank160 | Rating / score1,324.27 | Votes3,123 | ObservationsN/A | Result date |
| Arenatext style control | Categoryrussian | Rank160 | Rating / score1,313.90 | Votes7,673 | ObservationsN/A | Result date |
| Arenatext style control | Categoryexclude ties | Rank161 | Rating / score1,285.70 | Votes27,059 | ObservationsN/A | Result date |
| Arenatext style control | Categoryoverall | Rank161 | Rating / score1,334.92 | Votes41,375 | ObservationsN/A | Result date |
Llama 3.1 405B Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- From $1.95 input, $1.95 output per 1M via Cortecs
- Tracked offerings
- 1
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderCortecs | Provider model IDllama-3.1-405b-instruct | Regionglobal | Input / 1M$1.95 | Output / 1M$1.95 | Context128K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Llama 3.1 405B Runtime Performance
Provider-specific output speed and catalog latency for Llama 3.1 405B Instruct. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderGoogle | Output Speed42.00 tok/s | Catalog Latency0.40 s | Max Input128K | Max Output128K | Updated |
| ProviderDeepInfra | Output Speed27.00 tok/s | Catalog Latency0.50 s | Max Input128K | Max Output128K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Llama 3.1 405B Specifications
Technical details for the model's default version.
- Version
- Llama 3.1 405B Instruct
- Released
- Jul 23, 2024
- Knowledge cutoff
- Unknown
- Parameters
- 405B
- Context window
- 128K
- Max output
- 128K
- Inputs
- text
- Outputs
- text
- Open weights
- Yes
- License
- Llama 3.1 Community License
Llama 3.1 405B 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 3.1 405B Instruct | Released | LLMBoard22.88 | Parameters405B | Context128K | Max output128K | Open weightsYes | LicenseLlama 3.1 Community License |
Models similar to Llama 3.1 405B
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
ME
Llama 4 Maverick
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 3.1 405B?
Key information about Llama 3.1 405B and its available data.
1 405B Instruct is a large language model from Meta that supports 8 languages and has a 128K token context length.
Data as of 2026-09-08.
FAQ
Common questions about Llama 3.1 405B.
When was Llama 3.1 405B released?
Llama 3.1 405B's default version was released on Jul 23, 2024.
How much does Llama 3.1 405B cost?
No official standard PAYG price is currently available for Llama 3.1 405B. The lowest tracked third-party offer starts at $1.95 input and $1.95 output via Cortecs.
Who created Llama 3.1 405B?
Llama 3.1 405B was created by Meta.
What is the context window for Llama 3.1 405B?
The default version has a 128K token context window.
Is Llama 3.1 405B open weight?
Yes. The default version is marked as open weight under Llama 3.1 Community License.
How many API providers offer Llama 3.1 405B?
1 provider offerings are linked to the default version.