Llama 3.1 70B Instruct
Summary
Llama 3.1 70B Instruct is an open-weight language model from Meta, released on 23 Jul 2024. It accepts text and generates text, with a 128K-token context window and up to 8K output tokens per response. As of 18 Nov 2026, the lowest listed API price is $0.72 per million input tokens and $0.72 per million output tokens (OpenRouter), across 2 providers we track. It is scheduled for retirement on 3 Nov 2026.
Open Llama instruction model for multilingual chat, reasoning, and coding
Last verified 18 Nov 2026. Every figure on this page links to its source.
Specifications
| Developer | Meta |
|---|---|
| Release date | 23 Jul 2024 [models.dev] |
| Status | Generally available [models.dev] |
| Weights | Open weights [Hugging Face] |
| License | llama3.1 [Hugging Face] |
| Parameters | 70.6B [Hugging Face] |
| Context window | 128K (128,000 tokens) [models.dev] |
| Max output | 8K tokens [models.dev] |
| Knowledge cutoff | Dec 2023 [models.dev] |
| Input | Text [models.dev] |
| Output | Text [models.dev] |
| Reasoning mode | No [models.dev] |
| Tool calling | Yes [models.dev] |
| Structured output | No [models.dev] |
| Retirement date | 3 Nov 2026 [OpenRouter] |
| Hugging Face | meta-llama/Meta-Llama-3.1-70B-Instruct [Hugging Face] |
Llama 3.1 70B Instruct API pricing by provider
USD per million tokens. Sorted by input price.
| Provider | Input | Output | Cached input | Context | Source |
|---|---|---|---|---|---|
| OpenRouter meta-llama/llama-3.1-70b-instruct | $0.72 | $0.72 | – | 131K | OpenRouter, 18 Nov 2026 |
| Vercel AI Gateway meta/llama-3.1-70b | $0.72 | $0.72 | – | 128K | models.dev, 18 Nov 2026 |
Running Llama 3.1 70B Instruct locally
Estimated memory, from the parameter count on Hugging Face. Includes about 0.8 GB of runtime overhead.
| Quantization | Weights | Total, 8K context | Total, 32K context |
|---|---|---|---|
| Q4_K_M (4-bit) | 43 GB | 44 GB | – |
| Q8_0 (8-bit) | 75 GB | 76 GB | – |
| FP16 / BF16 (16-bit) | 141 GB | 142 GB | – |
At 4-bit with an 8K conversation, Llama 3.1 70B Instruct needs about 44 GB, more than any single consumer graphics card; it needs a unified-memory machine, a multi-GPU setup or a split between graphics card and system RAM. The model's architecture is not public, so the conversation cache is not included. Check your own hardware, or see Best computer for local AI on our sister site AIPCs.
Frequently asked questions
What is the context window of Llama 3.1 70B Instruct?
Llama 3.1 70B Instruct has a context window of 128,000 tokens (128K) and can generate up to 8,192 tokens in a single response, according to models.dev.
How much does the Llama 3.1 70B Instruct API cost?
Llama 3.1 70B Instruct is listed from $0.72 per million input tokens and $0.72 per million output tokens (OpenRouter). Prices last verified 18 Nov 2026.
Is Llama 3.1 70B Instruct open source?
Llama 3.1 70B Instruct is an open-weight model: its weights are published on Hugging Face as meta-llama/Meta-Llama-3.1-70B-Instruct under the llama3.1 license. Check the license terms before commercial use.
When was Llama 3.1 70B Instruct released?
Meta released Llama 3.1 70B Instruct on 23 Jul 2024, according to models.dev.
Which providers offer Llama 3.1 70B Instruct?
We track Llama 3.1 70B Instruct on 2 providers: OpenRouter, Vercel AI Gateway.
How much VRAM does Llama 3.1 70B Instruct need?
About 44 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 76 GB at 8-bit. That is more than any single consumer graphics card holds. These are estimates from the model's published parameter count.
Similar models from other labs
Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.
| Model | Lab | Released | Context | Input $/1M | Output $/1M | Weights |
|---|---|---|---|---|---|---|
| Llama 3.1 Euryale 70B v2.2 | Sao10k | 28 Aug 2024 | 131K | $0.85 | $0.85 | Open |
| Command R | Cohere | 30 Aug 2024 | 128K | $0.15 | $0.60 | Open |
| Aya Expanse 32B | Cohere | 24 Oct 2024 | 128K | $0.50 | $1.50 | Open |
| Qwen2.5-VL 72B Instruct | Alibaba (Qwen) | 26 Jan 2025 | 128K | $0.80 | $1 | Open |
| WizardLM 2 8x22B | Microsoft | 15 Apr 2024 | 66K | $0.62 | $0.62 | Open |
| Mistral Large 2407 | Mistral AI | 24 Jul 2024 | 128K | $2 | $6 | Open |
Change history
- Name changed from Llama-3.1-70B-Instruct to Llama 3.1 70B Instruct
- Context window changed from 131K to 128K
- Context window changed from 128K to 131K
- Name changed from Llama 3.1 70B Instruct to Llama-3.1-70B-Instruct
- Context window changed from 131K to 128K
- Tool calling changed from no to yes
- Name changed from Llama-3.1-70B-Instruct to Llama 3.1 70B Instruct
- Context window changed from 128K to 131K
- Tool calling changed from yes to no
- Name changed from Llama 3.1 70B Instruct to Llama-3.1-70B-Instruct
- Context window changed from 131K to 128K
- Tool calling changed from no to yes
- Name changed from Llama-3.1-70B-Instruct to Llama 3.1 70B Instruct
- Context window changed from 128K to 131K
- Tool calling changed from yes to no
- Name changed from Llama 3.1 70B Instruct to Llama-3.1-70B-Instruct
- Context window changed from 131K to 128K
- Tool calling changed from no to yes
- Name changed from Llama-3.1-70B-Instruct to Llama 3.1 70B Instruct
- Context window changed from 128K to 131K
- Name changed from Llama 3.1 70B Instruct to Llama-3.1-70B-Instruct
- Tool calling changed from yes to no
- Structured output changed from yes to no
- OpenRouter price changed: input $0.40 → $0.72, output $0.40 → $0.72 per 1M tokens
- Max output changed from 16K to 8K
Sources
- https://openrouter.ai/meta-llama/llama-3.1-70b-instruct
- https://github.com/sst/models.dev/blob/dev/providers/vercel/models/meta/llama-3.1-70b.toml
- https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct
- https://github.com/sst/models.dev/blob/dev/providers/openrouter/models/meta-llama/llama-3.1-70b-instruct.toml