LLM Pricing
PricingLeaderboardsToolsProvidersReleasesGuides

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list

Llama 3.3 70B

meta·meta/llama3-3-70b·GA·Open weights·llama series·NEW
Open Llama instruction model for multilingual chat, reasoning, and coding

Llama 3.3 70B by meta is offered by 6 providers on this page. Its reference price is $1.75 per 1M input tokens and $2.75 per 1M output tokens. The lowest paid channel is STACKIT at $0.53 / $0.76 per 1M, about 3.3× below the reference price.

The context window is 128,000 tokens at the reference host, but hosts report different limits, from 128,000 to 131,072, so the usable window depends on the provider you pick. It supports tool use. The weights are open, so it can also be self-hosted or served through a gateway of your choice. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use. Its training knowledge cuts off at 2023-12.

Specs & pricing

Input / output per 1M tokens
Reference price·NanoGPT
$1.75 / $2.75
Blended $2.00 · Cache read $1.75
Lowest paid·STACKITCloud
$0.53 / $0.76
Blended $0.588 · 3.3× spread
Context
128,000
Output limit
16,384
Knowledge cutoff
2023-12
Released / updated
2024-12-06 / 2026-07-02
Capabilities
Reasoning✓ Tool useStructured output✓ TemperatureAttachments
⚠ Providers report capability flags inconsistently
Modalities
Text

Available at 6 providers6 with public prices

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
STACKIT
cortecs/Llama-3.3-70B-Instruct-FP8-Dynamic
Cloud$0.53$0.76——128,0004,096
Groq
llama-3.3-70b-versatile
Cloud$0.59$0.79——131,072 ⚠32,768
Weights & Biases
meta-llama/Llama-3.3-70B-Instruct
Cloud$0.71$0.71$0.71—128,000128,000
Together AI
meta-llama/Llama-3.3-70B-Instruct-Turbo
Cloud$1.04$1.04——131,072 ⚠131,072
Venice AI
llama-3.3-70b
Gateway$0.70$2.80——128,0004,096
NanoGPTGateway$1.75$2.75$1.75—128,00016,384

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Your usage cost

1STACKIT$144.00
2Groq$157.50
3Weights & Biases$177.50
4Together AI$260.00
5Venice AI$280.00
6NanoGPT$487.50
The cheapest paid channel is the only channel.

Related models

Llama 4 Mavericksame series$0.20 / $0.696Llama Guard 4 12Bsame series$0.18 / $0.18Omega Directive 24B Unslop v2.0same series$0.50 / $0.50Qwen3-Coder 30B-A3B Instructcheaper alternative$0.216 / $0.861Qwen3-Next 80B-A3B Instructcheaper alternative$0.144 / $0.574Qwen3 VL 235B A22B Instructcheaper alternative$0.40 / $1.60Qwen3 Coder Nextcheaper alternative$0.50 / $1.20

Price historyone sample accumulated per data sync

Input list $1.75Output list $2.75Min blended $0.588

Price history accumulates from each data sync; currently only 1 sample(s) (2026-08-13). Each future sync adds a point, and once accumulated a line is drawn here.

Data partly from models.dev (MIT) · NanoGPT official docs ↗