InferIndex
Language, English, EN
Account

ModelsMetaLlama 3.3 70B Instruct

Llama 3.3 70B Instruct

meta-llama/llama-3.3-70b-instruct131k context Compare →
MedianThe median blended price among comparable offers: one price per provider (its cheapest route), standard tier, current prices only. Priority tiers and prices converted from points are not counted.
$0.72
Most expensiveThe highest blended price among comparable offers: one price per provider (its cheapest route), standard tier, current prices only. Priority tiers and prices converted from points are not counted.
$2.00
Offers
64
Providers
37

Cheapest right now

NanoGPT

The provider's published policy: prompts are not used for training.

131k

$0.095 blended / 1M tokens

No change7D
in
$0.05
out
$0.23
Cached input
–
Go to offer Go to offer at NanoGPT (opens in a new tab)via NanoGPT

Public leaderboards

  • Text Arena, style controlThe series the source publishes by default for text since 16 May 2025: ranks corrected for the style of the answers (style control). #253 of 419Best rank in the source's rank spread, 253rd to 278th, given the confidence intervals of all ranked models. 267th by score. 1,318 1,315 to 1,322Arena Score · 54,493 votes

Source: Arena (opens in a new tab) · Data under the CC BY 4.0 license (creativecommons.org/licenses/by/4.0) · CC BY 4.0 (opens in a new tab) · scores and ranks as published on 28 Oct 2026; the presentation is ours

Llama 3.3 70B Instruct costs from $0.095 per million tokens (at a 3:1 input:output mix): the cheapest of 27 comparable offers, at NanoGPT, checked on ; 64 offers are listed.

Listed offers, by price

Comparable offersOther offersSolid steps are comparable offers; dashed steps are the other offers, which are not counted in cheapest, median or most expensive.

Offers not counted in cheapest, median or most expensive (37)

  • $0.155 · DeepInfra via Eden AI · cannot be designated the cheapest
  • $0.155 · DeepInfra via Kilo Gateway · cannot be designated the cheapest
  • $0.155 · Glama via Glama · cannot be designated the cheapest
  • $0.155 · Martian via Martian · cannot be designated the cheapest
  • $0.155 · Nous Portal via Nous Portal · cannot be designated the cheapest
  • $0.155 · DeepInfra via OpenRouter · another route of the same provider
  • $0.165 · DeepInfra via Requesty · another route of the same provider
  • $0.174 · ShengSuanYun via ShengSuanYun · cannot be designated the cheapest
  • $0.198 · Nebius via LLM Gateway · cannot be designated the cheapest
  • $0.201 · Novita via Kilo Gateway · cannot be designated the cheapest
  • $0.201 · Novita via LLM Gateway · another route of the same provider
  • $0.201 · Novita via OpenRouter · another route of the same provider
  • $0.201 · Novita via Hugging Face · another route of the same provider
  • $0.201 · Novita via LLM API · cannot be designated the cheapest
  • $0.27 · FPT AI Factory · cannot be designated the cheapest
  • $0.28 · AkashML via Kilo Gateway · cannot be designated the cheapest
  • $0.29 · FlexAI via Eden AI · cannot be designated the cheapest
  • $0.29 · Parasail via Kilo Gateway · cannot be designated the cheapest
  • $0.39 · Novita via Requesty · another route of the same provider
  • $0.631 · io.net · cannot be designated the cheapest
  • $0.64 · Groq via Kilo Gateway · cannot be designated the cheapest
  • $0.70 · Arkanecloud via Eden AI · cannot be designated the cheapest
  • $0.71 · CoreWeave via Kilo Gateway · cannot be designated the cheapest
  • $0.72 · Amazon Bedrock via Eden AI · cannot be designated the cheapest
  • $0.72 · Google AI Studio via Kilo Gateway · cannot be designated the cheapest
  • $0.72 · Google AI Studio via Kilo Gateway · cannot be designated the cheapest
  • $0.726 · IONOS via Eden AI · cannot be designated the cheapest
  • $0.726 · Cortecs via Cortecs · cannot be designated the cheapest
  • $0.727 · Poe via Poe · price converted from points
  • $0.74 · OVHcloud via Eden AI · cannot be designated the cheapest
  • $0.75 · SambaNova · another route of the same provider
  • $0.75 · SambaNova via Kilo Gateway · cannot be designated the cheapest
  • $0.783 · Cloudflare via Kilo Gateway · cannot be designated the cheapest
  • $0.783 · Cloudflare via OpenRouter · another route of the same provider
  • $0.938 · Cerebras via LLM Gateway · cannot be designated the cheapest
  • $1.01 · Scaleway via Eden AI · cannot be designated the cheapest
  • $1.23 · Greenpt via Eden AI · cannot be designated the cheapest

Price historyThe cheapest offer each day, against the lab's official list price when it publishes one. Each day at 00:00 UTC, the offer the API would have selected as the cheapest, with the same exclusions as /cheapest. Blended price, promotions included.

Cheapest offer and official list price, by day, 30 days
DayCheapest offerOfficial list price
2026-10-02$0.095, NanoGPT–
2026-10-03$0.095, NanoGPT–
2026-10-04$0.095, NanoGPT–
2026-10-05$0.095, NanoGPT–
2026-10-06$0.095, NanoGPT–
2026-10-07$0.095, NanoGPT–
2026-10-08$0.095, NanoGPT–
2026-10-09$0.095, NanoGPT–
2026-10-10$0.095, NanoGPT–
2026-10-11$0.095, NanoGPT–
2026-10-12$0.095, NanoGPT–
2026-10-13$0.095, NanoGPT–
2026-10-14$0.095, NanoGPT–
2026-10-15$0.095, NanoGPT–
2026-10-16$0.095, NanoGPT–
2026-10-17$0.095, NanoGPT–
2026-10-18$0.095, NanoGPT–
2026-10-19$0.095, NanoGPT–
2026-10-20$0.095, NanoGPT–
2026-10-21$0.095, NanoGPT–
2026-10-22$0.095, NanoGPT–
2026-10-23$0.095, NanoGPT–
2026-10-24$0.095, NanoGPT–
2026-10-25$0.095, NanoGPT–
2026-10-26$0.095, NanoGPT–
2026-10-27$0.095, NanoGPT–
2026-10-28$0.095, NanoGPT–
2026-10-29$0.095, NanoGPT–
2026-10-30$0.095, NanoGPT–
2026-10-31$0.095, NanoGPT–

Cheapest offer

Offers

The 25 cheapest of 64 offers

Offers for Llama 3.3 70B Instruct, one row per offer, sorted by blended price.
RankProvider · viaYou buy access through an aggregator: one account and one API that forward your requests to other providers. "Via" names the aggregator; the price is the one it charges you. When "via" says "hosted API", no aggregator is involved: the model belongs to a lab and another provider sells it.Input $/1MOutput $/1MBlended $/1MA single price that mixes input and output: 3 input tokens for 1 output token, per million tokens. It makes offers easy to rank; your real cost depends on your own mix.PrecisionHow finely the model's numbers are stored on the provider's servers. Lower precision (fp8, fp4) is cheaper to run and can slightly change answer quality; "unknown" means the provider doesn't say. ContextThe maximum amount of text, counted in tokens, the model can take in and write in a single request. Past this limit, the request is refused or cut.ConditionsWhat a provider says about how it handles your data and who can sign up: training on prompts, processing region, sign-up requirements. Each comes from the provider's published documents; when nothing is published, it stays unknown.SignalsOffer
1 NanoGPT 0.05  0.23  0.095 – 131k
Go to offer at NanoGPT (opens in a new tab)
2 DeepInfra 0.10  0.32  0.155 fp8 131k United States
Go to offer at DeepInfra (opens in a new tab)
3 DeepInfravia Eden AI 0.10  0.32  0.155 fp8 131k United States Unverified. No public page for this offer: listed, not designated cheapest
: Eden AI (opens in a new tab). No public price page for this model
4 DeepInfravia Kilo Gateway 0.10  0.32  0.155 fp8 131k United States Unverified. No public page for this offer: listed, not designated cheapest
: Kilo Gateway (opens in a new tab). No public price page for this model
5 Glama 0.10  0.32  0.155 – 128k Unverified. No public page for this offer: listed, not designated cheapest
: Glama (opens in a new tab). No public price page for this model
6 Martian 0.10  0.32  0.155 – – worldwide Unverified. Neither seller nor price found on its page for 72 hours: listed, not designated cheapest
See pricing at Martian (opens in a new tab)
7 Nous Portal 0.10  0.32  0.155 – 131k Unverified. Neither seller nor price found on its page for 72 hours: listed, not designated cheapest
See pricing at Nous Portal (opens in a new tab)
8 DeepInfravia OpenRouter 0.10  0.32  0.155 fp8 131k United States Fee 5.5%: OpenRouter adds a 5.5% fee when you buy credits (read on openrouter.ai/pricing on Oct 30). Offers are ranked with this fee included.
Go to offer at OpenRouter (opens in a new tab)
9 DeepInfravia Requesty 0.12  0.30  0.165 fp8 131k United States Fee 5%: Requesty adds a 5% markup (read on requesty.ai/pricing on Oct 30). Offers are ranked with this fee included.
Go to offer at Requesty (opens in a new tab)
10 ShengSuanYun 0.126 0.316 0.174 – 128k Unverified. No public page for this offer: listed, not designated cheapest
: ShengSuanYun (opens in a new tab). No public price page for this model
11 InferenceNet 0.13  0.40  0.198 – 131k
Go to offer at InferenceNet (opens in a new tab)
12 Nebiusvia LLM Gateway 0.13  0.40  0.198 – – worldwide Fee 5%: LLM Gateway adds a 5% fee on credit usage, plus 1.5% for cards issued outside the US (read on llmgateway.io/pricing on Oct 30). Offers are ranked with the 5% included.
Go to offer at LLM Gateway (opens in a new tab)
13 Nebiusvia Requesty 0.13  0.40  0.198 – 128k worldwide Fee 5%: Requesty adds a 5% markup (read on requesty.ai/pricing on Oct 30). Offers are ranked with this fee included.
Go to offer at Requesty (opens in a new tab)
14 Novitavia Kilo Gateway 0.135 0.40  0.201 bf16 12k Unverified. No public page for this offer: listed, not designated cheapest
: Kilo Gateway (opens in a new tab). No public price page for this model
15 Novitavia LLM Gateway 0.135 0.40  0.201 bf16 – Fee 5%: LLM Gateway adds a 5% fee on credit usage, plus 1.5% for cards issued outside the US (read on llmgateway.io/pricing on Oct 30). Offers are ranked with the 5% included.
Go to offer at LLM Gateway (opens in a new tab)
16 Novitavia OpenRouter 0.135 0.40  0.201 bf16 12k Fee 5.5%: OpenRouter adds a 5.5% fee when you buy credits (read on openrouter.ai/pricing on Oct 30). Offers are ranked with this fee included.
Go to offer at OpenRouter (opens in a new tab)
17 Novita 0.135 0.40  0.201 bf16 12k
See pricing at Novita (opens in a new tab)
18 Novitavia Hugging Face 0.135 0.40  0.201 bf16 12k
See pricing at Hugging Face (opens in a new tab)
19 Novitavia LLM API 0.135 0.40  0.201 bf16 – Unconfirmed. Last checked 18 October 2026: the last published value, not a confirmed one Unverified. The source page shows one price per model, not this seller's: listed, not designated cheapest
See pricing at LLM API (opens in a new tab)
20 FPT AI Factory 0.21  0.45  0.27  – 32k Unverified. No public page for this offer: listed, not designated cheapest
: FPT AI Factory (opens in a new tab). No public price page for this model
21 DeepInfravia Requesty 0.23  0.40  0.273 fp8 131k United States Fee 5%: Requesty adds a 5% markup (read on requesty.ai/pricing on Oct 30). Offers are ranked with this fee included.
Go to offer at Requesty (opens in a new tab)
22 AkashMLvia Kilo Gateway 0.20  0.52  0.28  fp8 131k Unverified. No public page for this offer: listed, not designated cheapest
: Kilo Gateway (opens in a new tab). No public price page for this model
23 AkashMLvia OpenRouter 0.20  0.52  0.28  fp8 131k Fee 5.5%: OpenRouter adds a 5.5% fee when you buy credits (read on openrouter.ai/pricing on Oct 30). Offers are ranked with this fee included.
Go to offer at OpenRouter (opens in a new tab)
24 FlexAIvia Eden AI 0.22  0.50  0.29  fp8 131k Unverified. No public page for this offer: listed, not designated cheapest
: Eden AI (opens in a new tab). No public price page for this model
25 Parasailvia Kilo Gateway 0.22  0.50  0.29  fp8 131k worldwide Unverified. No public page for this offer: listed, not designated cheapest
: Kilo Gateway (opens in a new tab). No public price page for this model

No training on promptsPrompts may be used for trainingTraining terms not published

backend not named. The router bills this price itself without naming who serves the model.

  • NanoGPT

    $0.095

    NanoGPT

    in
    $0.05 
    out
    $0.23 
    Precision
    –
    Context
    131k

    No training on prompts

    Go to offer Go to offer at NanoGPT (opens in a new tab)via NanoGPT

  • DeepInfra

    $0.155

    direct

    in
    $0.10 
    out
    $0.32 
    Precision
    fp8
    Context
    131k

    No training on promptsUnited States

    Go to offer Go to offer at DeepInfra (opens in a new tab)at DeepInfra

  • DeepInfra

    $0.155

    Eden AI

    in
    $0.10 
    out
    $0.32 
    Precision
    fp8
    Context
    131k

    Prompts may be used for trainingUnited StatesUnverified. No public page for this offer: listed, not designated cheapest

    Seller's site: Eden AI (opens in a new tab)No public price page for this model

  • DeepInfra

    $0.155

    Kilo Gateway

    in
    $0.10 
    out
    $0.32 
    Precision
    fp8
    Context
    131k

    Training terms not publishedUnited StatesUnverified. No public page for this offer: listed, not designated cheapest

    Seller's site: Kilo Gateway (opens in a new tab)No public price page for this model

  • Glama

    $0.155

    Glama

    in
    $0.10 
    out
    $0.32 
    Precision
    –
    Context
    128k

    No training on promptsUnverified. No public page for this offer: listed, not designated cheapestbackend not named: The router bills this price itself without naming who serves the model.

    Seller's site: Glama (opens in a new tab)No public price page for this model

  • Martian

    $0.155

    Martian

    in
    $0.10 
    out
    $0.32 
    Precision
    –
    Context
    –

    Training terms not publishedworldwideUnverified. Neither seller nor price found on its page for 72 hours: listed, not designated cheapestbackend not named: The router bills this price itself without naming who serves the model.

    See pricing See pricing at Martian (opens in a new tab)via Martian

  • Nous Portal

    $0.155

    Nous Portal

    in
    $0.10 
    out
    $0.32 
    Precision
    –
    Context
    131k

    Prompts may be used for trainingUnverified. Neither seller nor price found on its page for 72 hours: listed, not designated cheapestbackend not named: The router bills this price itself without naming who serves the model.

    See pricing See pricing at Nous Portal (opens in a new tab)via Nous Portal

  • DeepInfra

    $0.155

    OpenRouter

    in
    $0.10 
    out
    $0.32 
    Precision
    fp8
    Context
    131k

    No training on promptsUnited StatesFee 5.5%: OpenRouter adds a 5.5% fee when you buy credits (read on openrouter.ai/pricing on Oct 30). Offers are ranked with this fee included.

    Go to offer Go to offer at OpenRouter (opens in a new tab)via OpenRouter

  • DeepInfra

    $0.165

    Requesty

    in
    $0.12 
    out
    $0.30 
    Precision
    fp8
    Context
    131k

    Prompts may be used for trainingUnited StatesFee 5%: Requesty adds a 5% markup (read on requesty.ai/pricing on Oct 30). Offers are ranked with this fee included.

    Go to offer Go to offer at Requesty (opens in a new tab)via Requesty

  • ShengSuanYun

    $0.174

    ShengSuanYun

    in
    $0.126
    out
    $0.316
    Precision
    –
    Context
    128k

    Training terms not publishedUnverified. No public page for this offer: listed, not designated cheapest

    Seller's site: ShengSuanYun (opens in a new tab)No public price page for this model

  • InferenceNet

    $0.198

    direct

    in
    $0.13 
    out
    $0.40 
    Precision
    –
    Context
    131k

    No training on prompts

    Go to offer Go to offer at InferenceNet (opens in a new tab)at InferenceNet

  • Nebius

    $0.198

    LLM Gateway

    in
    $0.13 
    out
    $0.40 
    Precision
    –
    Context
    –

    Training terms not publishedworldwideFee 5%: LLM Gateway adds a 5% fee on credit usage, plus 1.5% for cards issued outside the US (read on llmgateway.io/pricing on Oct 30). Offers are ranked with the 5% included.

    Go to offer Go to offer at LLM Gateway (opens in a new tab)via LLM Gateway

  • Nebius

    $0.198

    Requesty

    in
    $0.13 
    out
    $0.40 
    Precision
    –
    Context
    128k

    Prompts may be used for trainingworldwideFee 5%: Requesty adds a 5% markup (read on requesty.ai/pricing on Oct 30). Offers are ranked with this fee included.

    Go to offer Go to offer at Requesty (opens in a new tab)via Requesty

  • Novita

    $0.201

    Kilo Gateway

    in
    $0.135
    out
    $0.40 
    Precision
    bf16
    Context
    12k

    Training terms not publishedUnverified. No public page for this offer: listed, not designated cheapest

    Seller's site: Kilo Gateway (opens in a new tab)No public price page for this model

  • Novita

    $0.201

    LLM Gateway

    in
    $0.135
    out
    $0.40 
    Precision
    bf16
    Context
    –

    No training on promptsFee 5%: LLM Gateway adds a 5% fee on credit usage, plus 1.5% for cards issued outside the US (read on llmgateway.io/pricing on Oct 30). Offers are ranked with the 5% included.

    Go to offer Go to offer at LLM Gateway (opens in a new tab)via LLM Gateway

  • Novita

    $0.201

    OpenRouter

    in
    $0.135
    out
    $0.40 
    Precision
    bf16
    Context
    12k

    No training on promptsFee 5.5%: OpenRouter adds a 5.5% fee when you buy credits (read on openrouter.ai/pricing on Oct 30). Offers are ranked with this fee included.

    Go to offer Go to offer at OpenRouter (opens in a new tab)via OpenRouter

  • Novita

    $0.201

    direct

    in
    $0.135
    out
    $0.40 
    Precision
    bf16
    Context
    12k

    No training on prompts

    See pricing See pricing at Novita (opens in a new tab)at Novita

  • Novita

    $0.201

    Hugging Face

    in
    $0.135
    out
    $0.40 
    Precision
    bf16
    Context
    12k

    No training on prompts

    See pricing See pricing at Hugging Face (opens in a new tab)via Hugging Face

  • Novita

    $0.201

    LLM API

    in
    $0.135
    out
    $0.40 
    Precision
    bf16
    Context
    –

    No training on promptsUnconfirmed. Last checked 18 October 2026: the last published value, not a confirmed oneUnverified. The source page shows one price per model, not this seller's: listed, not designated cheapest

    See pricing See pricing at LLM API (opens in a new tab)via LLM API

  • FPT AI Factory

    $0.27

    direct

    in
    $0.21 
    out
    $0.45 
    Precision
    –
    Context
    32k

    Training terms not publishedUnverified. No public page for this offer: listed, not designated cheapest

    Seller's site: FPT AI Factory (opens in a new tab)No public price page for this model

  • DeepInfra

    $0.273

    Requesty

    in
    $0.23 
    out
    $0.40 
    Precision
    fp8
    Context
    131k

    Prompts may be used for trainingUnited StatesFee 5%: Requesty adds a 5% markup (read on requesty.ai/pricing on Oct 30). Offers are ranked with this fee included.

    Go to offer Go to offer at Requesty (opens in a new tab)via Requesty

  • AkashML

    $0.28

    Kilo Gateway

    in
    $0.20 
    out
    $0.52 
    Precision
    fp8
    Context
    131k

    Training terms not publishedUnverified. No public page for this offer: listed, not designated cheapest

    Seller's site: Kilo Gateway (opens in a new tab)No public price page for this model

  • AkashML

    $0.28

    OpenRouter

    in
    $0.20 
    out
    $0.52 
    Precision
    fp8
    Context
    131k

    Training terms not publishedFee 5.5%: OpenRouter adds a 5.5% fee when you buy credits (read on openrouter.ai/pricing on Oct 30). Offers are ranked with this fee included.

    Go to offer Go to offer at OpenRouter (opens in a new tab)via OpenRouter

  • FlexAI

    $0.29

    Eden AI

    in
    $0.22 
    out
    $0.50 
    Precision
    fp8
    Context
    131k

    Prompts may be used for trainingUnverified. No public page for this offer: listed, not designated cheapest

    Seller's site: Eden AI (opens in a new tab)No public price page for this model

  • Parasail

    $0.29

    Kilo Gateway

    in
    $0.22 
    out
    $0.50 
    Precision
    fp8
    Context
    131k

    Training terms not publishedworldwideUnverified. No public page for this offer: listed, not designated cheapest

    Seller's site: Kilo Gateway (opens in a new tab)No public price page for this model
Show all 64 offers

2 flex or batch offers hidden (lower priority, variable latency).

Blended = (3 × input + output) / 4, per million tokens. Prices in USD, converted at the ECB rate when the provider bills in another currency.

Method

Monthly cost for your workload

Computed by the API for the three cheapest offers, per 30-day month.
  1. NanoGPT$6.45per month
  2. DeepInfra$10.80per month+$4.35 vs the cheapest
  3. InferenceNet$13.80per month+$7.35 vs the cheapest

Host it yourself instead? See the break-even

Compare Llama 3.3 70B Instruct with

Related models

More from Meta

All Meta models

Same index family

Use this data

The index is queryable. The API is free to use.
Free API
curl https://api.inferindex.dev/cheapest?model=meta-llama/llama-3.3-70b-instruct
MCP server
claude mcp add --transport http inferindex https://mcp.inferindex.dev/mcp

Also works in Cursor and Codex: see the MCP page.