This model is deprecated. We only continue performance benchmarking for the default 10k input token workload. Results for other workloads are historical and no longer updated.

Meta has launched a newer model, Llama 3.1 70B. We suggest considering it instead.

For more information, see comparison of Llama 3.1 70B to other models and API provider benchmarks for Llama 3.1 70B.

Llama 3 Instruct 70B logo
Meta
•

Open weights model

•

Released April 2024

Llama 3 Instruct 70B API 服务商基准测试与分析

模型比较

分析 Llama 3 Instruct 70B 各 API 服务商的性能指标，包括延迟（首 Token 延迟）、输出速度（每秒输出 token 数）、价格等。参与基准测试的 API 服务商包括 Novita、Amazon Bedrock和Replicate。

速度最快

输出速度

共 3 家服务商

延迟最低

首 Token 延迟

共 3 家服务商

价格最低

#1
NovitaNovita
$0.53
#2
ReplicateReplicate
$0.86
#3
AmazonAmazon
$2.74

每 100 万 token 的混合价格

共 3 家服务商

可通过 3 家 API 服务商使用 Llama 3 70B，每家服务商的性能特征和价格各不相同。下方对比了各服务商的关键指标。

  • 价格方面，Novita（0.53）、Replicate（0.86）和Amazon（2.74） 每 100 万 token 的混合价格最低。

该模型暂无服务商基准测试数据。
有关模型详情及其与其他模型的智能比较，请参阅 Llama 3 Instruct 70B 的模型页面。

亮点

速度

Output tokens per second · Higher is better
No data available

端到端响应时间

Seconds · Lower is better
No data available

价格

USD per 1M tokens (blended) · Lower is better

价格

Pricing: Cache Hit, Input, and Output

Price (USD per M Tokens) · Lower is better · 10,000 input tokens

Price per token for cached prompts (previously processed), typically offering a significant discount compared to regular input price, represented as USD per million tokens. The values shown here are the cache hit price; cache write and cache storage are billed separately and vary by provider — see "Cache pricing by provider" for detail.

Pricing: Blended Price

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended) · Lower is better

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

Output Speed vs. Price

Blended at 7:2:1 (cache-input-output) · Output speed: output tokens per second · USD per 1M tokens (blended)
Most attractive quadrant
No data available

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

速度

按输出速度（每秒 token 数）衡量

Output Speed: Llama 3 Instruct 70B

Output speed: output tokens per second · 10,000 input tokens
No data available

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).

Latency vs. Output Speed

Latency: seconds to first token received · Output speed: output tokens per second · 10,000 input tokens
Most attractive quadrant
No data available

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

延迟

按首 Token 延迟（秒）衡量

首 Token 延迟

Seconds to first token received · Lower is better · 10,000 input tokens
No data available

Time to first token received, in seconds, after API request sent. For reasoning models which share reasoning tokens, this will be the first reasoning token. For models which do not support streaming, this represents time to receive the completion.

Figures represent median (P50) measurement over the past 72 hours to reflect sustained changes in performance.

端到端响应时间

Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed

端到端响应时间

Seconds to output 500 tokens, including reasoning model 'thinking' time · Lower is better · 10,000 input tokens
No data available

Seconds to receive a 500 token response. Key components:

  • Input time: Time to receive the first response token
  • Thinking time (only for reasoning models): Time reasoning models spend outputting tokens to reason prior to providing an answer. Amount of tokens based on the average reasoning tokens across a diverse set of 60 prompts (methodology details).
  • Answer time: Time to generate 500 output tokens, based on output speed

关键比较指标与 API 功能

Novita 标志Novita
8.19k
开放
--
--
--
--
--
Amazon Bedrock 标志Amazon Bedrock
8.19k
开放
--
--
--
--
--
Replicate 标志Replicate
8.19k
开放
--
--
--
--
--

常见问题

关于 Llama 3 Instruct 70B 服务商的常见问题

可通过 3 家 API 服务商使用 Llama 3 Instruct 70B：Novita、Amazon和Replicate。每家服务商的性能特征和价格各不相同。

目前可通过我们进行基准测试和跟踪的 3 家 API 服务商使用 Llama 3 Instruct 70B。

按混合价格计算，Llama 3 Instruct 70B 最实惠的服务商是 Novita（每 100 万 token $0.53）、Replicate（每 100 万 token $0.86）和Amazon（每 100 万 token $2.74）。混合价格采用缓存命中/输入/输出 token 为 7:2:1 的比例。

Llama 3 Instruct 70B 输入 token 价格最低的服务商是 Novita（每 100 万输入 token $0.51）、Replicate（每 100 万输入 token $0.65）和Amazon（每 100 万输入 token $2.65）。

Llama 3 Instruct 70B 输出 token 价格最低的服务商是 Novita（每 100 万输出 token $0.74）、Replicate（每 100 万输出 token $2.75）和Amazon（每 100 万输出 token $3.50）。

Llama 3 Instruct 70B 各服务商价格最多相差 5.1 倍。最实惠的是 Novita，每 100 万 token 收费 $0.53；Amazon 则收费 $2.74。

提供 Llama 3 Instruct 70B 的 3 家服务商中，有 1 家支持 JSON 模式：Novita。

选择 Llama 3 Instruct 70B 服务商时，请考虑：输出速度（适合吞吐量密集型任务）、延迟（适合需要快速首次响应的交互式应用）、价格（适合成本敏感型工作负载），以及 JSON 模式或函数调用等 API 功能。

会。服务商性能可能因基础设施变化、负载均衡和更新而随时间波动。我们持续对所有服务商进行基准测试，并在“随时间变化”图表中展示历史性能趋势。

有关 Llama 3 Instruct 70B 的智能、能力、模态及其与其他模型的比较，请参阅模型概览页面。 查看模型概览