Model Library
MetaReleased December 6, 2024Llama 3.3 70B
Meta · Released December 6, 2024
Intelligence score
9
At a glance
- Intelligence
- 9
- Speed
- 85.5tokens/sec
- Latency
- 1.59stime to first token
- Input price
- $0.58per 1M tokens
- Output price
- $0.71per 1M tokens
- Context window
- 130Ktokens
- Parameters
- 70billions
Summary
Comparison summary
Llama 3.3 Instruct 70B scores 9 on the Artificial Analysis Intelligence Index, placing it above average among other open weight non-reasoning models of similar size (median: 7). Llama 3.3 Instruct 70B generates output at 85.5 tokens per second (based on the median across providers serving the model), which is above average compared to other open weight non-reasoning models of similar size (median: 81.5 t/s). Llama 3.3 Instruct 70B costs $0.58 per 1M input tokens (somewhat higher than average, median: $0.53) and $0.71 per 1M output tokens (better than average, median: $1.05), based on the median across providers serving the model.
Specifications
Technical specifications
| Model name | Llama 3.3 70B |
|---|---|
| Provider | Meta |
| Release date | December 6, 2024 |
| Intelligence score | 9 |
| Speed | 85.5 tokens/sec |
| Latency (TTFT) | 1.59s |
| Input price | $0.58 / 1M tokens |
| Output price | $0.71 / 1M tokens |
| Context window | 130,000 tokens |
| Parameters | 70B |
| Reasoning | No |
| Input modalities | text |
| Output modalities | — |
| Open weights | Yes |
| License | LLAMA 3 |
| API availability | — |
| Source | Model Intelligence legacy profile: llama-3-3-instruct-70b |
Pricing
Token prices
Price per 1M tokens
Llama 3.3 70B, as listed in the catalog record
- Input$0.58
- Output$0.71
| Token type | Price per 1M |
|---|---|
| Input | $0.58 |
| Output | $0.71 |
Model Intelligence legacy profile: llama-3-3-instruct-70b
FAQ · 21
Frequently asked questions
Llama 3.3 Instruct 70B was released on December 6, 2024.
Llama 3.3 Instruct 70B was created by Meta.
Llama 3.3 Instruct 70B scores 9 on the Artificial Analysis Intelligence Index, placing it above average among other open weight non-reasoning models of similar size (median: 7).
Llama 3.3 Instruct 70B generates output at 85.5 tokens per second (based on the median across providers serving the model), which is above average compared to other open weight non-reasoning models of similar size (median: 81.5 t/s).
Llama 3.3 Instruct 70B has a time to first token (TTFT) of 1.59s (based on the median across providers serving the model), which is better than average compared to other open weight non-reasoning models of similar size (median: 1.59s).
Llama 3.3 Instruct 70B costs $0.58 per 1M input tokens (somewhat higher than average, median: $0.53) and $0.71 per 1M output tokens (better than average, median: $1.05), based on the median across providers serving the model.
Llama 3.3 Instruct 70B costs $0.58 per 1M input tokens and $0.71 per 1M output tokens (based on the median across providers serving the model). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.59 per 1M tokens. Pricing may vary by provider.
When evaluated on the Intelligence Index, Llama 3.3 Instruct 70B generated 3.8M output tokens, which is very competitive compared to other open weight non-reasoning models of similar size (median: 9.2M).
No, Llama 3.3 Instruct 70B is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
Llama 3.3 Instruct 70B supports text only input.
Llama 3.3 Instruct 70B supports text only output.
No, Llama 3.3 Instruct 70B does not support image input. It can only process text.
No, Llama 3.3 Instruct 70B is not multimodal. It only supports text only input.
Llama 3.3 Instruct 70B has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Llama 3.3 Instruct 70B is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Llama 3.3 Instruct 70B has 70 billion parameters.
Llama 3.3 Instruct 70B is released under the LLAMA 3.3 COMMUNITY LICENSE AGREEMENT license. This license allows commercial use.
Llama 3.3 Instruct 70B achieves a score of 9 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Llama 3.3 Instruct 70B has a knowledge cutoff of December 2023. The model's training data includes information up to this date.
Yes, Llama 3.3 Instruct 70B is available via API through 19 providers.
Llama 3.3 Instruct 70B is available through 19 API providers.