Llama 3.1 405B
Flagship open Llama — 405B parameters, comparable to GPT-4 on many tasks.
- Input
- $3.50 / 1M tokens
- Output
- $3.50 / 1M tokens
- Cached read
- — / 1M tokens
- Cached write
- — / 1M tokens
- Batch discount
- —%
- Source
- Llama 3.1 405B pricing
- Verified
- Apr 5, 2026 (High)
Capabilities
- Modalities
- text→text
- Capabilities
- batchSupportpromptCachingfunctionCallingstructuredOutputs
- Strengths
- Largest open-weight model, GPT-4-class quality
- Tradeoffs
- Needs multi-GPU, expensive to host
Official Links
Benchmark Coverage
| Benchmark | Version | Score | Date | Source | Notes |
|---|---|---|---|---|---|
| MMLU | 2024 | 87.3 % | Jul 1, 2024 | Meta | Vendor-reported |
| HumanEval | 2024 | 89 % | Jul 1, 2024 | Meta | Vendor-reported |
Release History
| Release | Alias | Lifecycle | Release Date | Deprecation | Shutdown | Summary |
|---|---|---|---|---|---|---|
| Llama 3.1 405B | llama-3-1-405b | Active | Jul 23, 2024 | — | — | Current published model family snapshot. |
Host Coverage
| Host | Type | Context | Pricing Note | Differences |
|---|---|---|---|---|
| Together AI | cloud | 128.0K | $3.50/$3.50 per MTok. | — |
| DeepInfra | cloud | 128.0K | $1.79/$1.79 per MTok. | — |
Migration Guidance
Use Llama 4 Maverick for better quality with cheaper MoE hosting.
Replacement models: llama-4-maverick
Change Events
| Date | Type | Title | Description | Source |
|---|---|---|---|---|
| Jul 23, 2024 | family_added | Llama 3.1 405B published | Initial public model family launch. | Llama 3.1 405B release notes |
Other models from Meta AI
Llama 3.1 70B, Llama 3.1 8B, Llama 3.2 11B Vision, Llama 3.2 90B Vision, Llama 3.3 70B, Llama 4 Maverick, Llama 4 Scout