Providers

API providers, route pricing, availability, and recent reliability signals.

Performance

Latency, throughput, and reliability signals from recent traffic.

Pricing

Effective prices over the last 30 days, with current provider list prices for context.

Benchmarks

Activity

Daily gateway activity over the last 30 days, with current UTC-day pace projection.

Apps Using This Model

Public apps observed in gateway usage for this model.

Model Uptime

Uptime trend for this model over the last 24 hours.

Quickstart

Start calling this model with endpoint-specific examples.

About

Key dates, capabilities, and model metadata.

Subscriptions

Commercial plans and bundled access that currently include this model.

Llama 3.3 70B Instruct Pricing, Benchmarks & Providers | Phaseo

Meta: Llama 3.3 70B Instruct

Chat Compare

Llama 3.3 70B Instruct is a language model from Meta for chat, writing, analysis, and general-purpose automation. It accepts text inputs and produces text outputs.

Providers

API providers, route pricing, availability, and recent reliability signals.


DeepInfra	$0.1	$0.32	--	--	--

AkashML	$0.13	$0.4	--	--	--

Nebius Token Factory Standard/Priority	$0.13	$0.4	--	--	--

NovitaAI	$0.135	$0.4	--	--	--

Groq	$0.59	$0.79	--	--	--

Weights & Biases	$0.71	$0.71	--	--	--

Together	$0.88	$0.88	--	--	--

Venice	$0.7	$2.8	--	--	--

Cloudflare	$0.293	$2.253	--	--	--

Fireworks	--	$0.9	--	--	--

DigitalOcean	$0.65	$0.65	--	--	--

Friendli	$0.6	$0.6	--	--	--

OVHcloud	$0.67	$0.67	--	--	--

Scaleway	$0.9	$0.9	--	--	--

Pricing

Weighted provider pricing over the last 30 days, with recent route pricing history below.

Effective pricing

Weighted by routed usage over the last 30 days; external and non-routable providers are excluded.

Weighted input price

Per 1M tokens

Weighted output price

Per 1M tokens


AkashML	$0.13	$0.4	--
DeepInfra	$0.1	$0.32	--
Groq	$0.59	$0.79	--
Nebius Token Factory	$0.13	$0.4	--
Nebius Token Factory (Fast)	$0.25	$0.75	--
NovitaAI	$0.135	$0.4	--
Together	$0.88	$0.88	--
Venice	$0.7	$2.8	--
Weights & Biases	$0.71	$0.71	--

No 7-day effective pricing is available for the selected service tier.

Quickstart

Start calling this model with endpoint-specific examples.

Get an API key

Create an API key inSettingsKeysand store it asPHASEO_API_KEY

Keep it server-side, never commit it, and rotate it immediately if exposed.

Send the request

Choose a supported endpoint, pick a main language, then select the example style you want to copy.

Supported endpoints

Supported API reference routes for this model.

POST

/v1/responses

Responses API reference

POST

/v1/chat/completions

Chat Completions API reference

POST

/v1/messages

Messages API reference

Streaming

import Phaseo from '@phaseo/sdk';

const client = new Phaseo({
  apiKey: process.env.PHASEO_API_KEY,
});

const response = await client.generateResponse({
    "model": "meta/llama-3.3-70b",
    "input": "Give me one fun fact about cURL."
});

const outputText = response.output
  ?.flatMap((item) => item.content ?? [])
  .find((item) => item.type === "output_text")
  ?.text;

console.log(outputText ?? response);

Accepted IDsClick to use and copy

Parameters

Aggregated across active providers for the responses route.

Routing will select a compatible provider when a parameter narrows availability, so this list stays model-facing instead of provider-facing.

View all parameters

Parameter	Description
`temperature`	Controls how random token selection can be.
`top_p`	Applies nucleus sampling by limiting candidates to a probability mass threshold.
`top_k`	Restricts sampling to the top-k candidate tokens on providers that expose it.
`max_tokens`	Caps output length on endpoints and providers that use the max_tokens field name.
`frequency_penalty`	Discourages repeated tokens in proportion to how often they already appeared.
`presence_penalty`	Encourages the model to explore new wording or topics after they first appear.
`repetition_penalty`	Applies provider-specific anti-repetition behavior outside the classic penalty fields.
`seed`	Requests deterministic sampling when the upstream provider supports seeded generation.
`stop`	Defines one or more sequences that terminate generation early.
`logprobs`	Requests token-level probability data in the response.
`tools`	Defines callable tools or functions the model can invoke.
`logit_bias`	Adjusts token selection bias directly when a provider exposes logit control.
`min_p`	Narrows sampling by discarding tokens below a minimum probability threshold.
`top_logprobs`	Limits how many alternative token probabilities are returned per position.

Docs:TypeScript SDK Responses API

Providers

Performance

Pricing

Benchmarks

Activity

Apps Using This Model

Model Uptime

Quickstart

About

Subscriptions

Providers

Performance

Pricing

Benchmarks

Activity

Apps Using This Model

Model Uptime

Quickstart

About

Subscriptions

Meta: Llama 3.3 70B Instruct

Providers

Performance