Models/Meta/Llama 3.3 70B Instruct

Llama 3.3 70B Instruct

Name: Llama 3.3 70B Instruct API access
Brand: Meta

Meta · Open weights · Released Dec 6, 2024 · meta-llama/llama-3.3-70b-instruct

Best $0.10 in / $0.32 out via DeepInfra · fp8

Compare Estimate cost

131K contextToolsStructured outputPrompt cachingOpen weights

Available from 11 providersUSD / MTOK · PRICES 23 MINUTES AGO · SPEED 2 MINUTES AGO

EST. MONTHLY COSTclick a row for detail · click headers to sort

PROVIDER	INPUT	OUTPUT	CACHE	CONTEXT	QUANT	TTFT	TPS	SMOKE	UPTIME
DeepInfra	$0.10	$0.32	—	131K	fp8	1.7s	39 tok/s	5/5	99.9%
Novita	$0.135	$0.40	—	12K	bf16	✕	✕	5/5	97.2%
AkashML	$0.20	$0.52	$0.10	131K	fp8	2.9s	46 tok/s	5/5	99.6%
Parasail	$0.22	$0.50	$0.11	131K	fp8	✕	✕	4/5	99.9%
Cloudflare	$0.293	$2.25	—	24K	fp8	4.6s	39 tok/s	4/5	99.9%
SambaNova	$0.45	$0.90	—	131K	—	✕	✕	—	99.9%
Groq	$0.59	$0.79	$0.295	131K	—	0.81s	250 tok/s	—	100.0%
CoreWeave	$0.71	$0.71	$0.71	128K	fp16	0.61s	83 tok/s	—	99.8%
Googlegoogle-vertex/us-central1	$0.72	$0.72	—	128K	—	✕	✕	—	—
Googlegoogle-vertex	$0.72	$0.72	—	128K	—	—	—	—	—
Together	$1.04	$1.04	—	131K	—	1.1s	33 tok/s	—	98.0%

TTFT/TPS: median (P50) over trailing 504h · chat10k profile (~10K in / 200 out) · measured via OpenRouter routing · ✕ = probed, endpoint returned no measurable stream · SMOKE: of 5 programmatic capability probes passed (JSON schema, instruction following, tool call, long-context, benign compliance; retry-once) · screened via OpenRouter routing · v0.2 · methodology v1.1 · measured 2 minutes ago

SERVE THIS MODEL AND NOT LISTED? LIST YOUR INFERENCE →

Recent changesRSS ↗

Oct 5, 2026endpoint removed · primeintellect12 → 11

Oct 4, 2026endpoint added · primeintellect11 → 12

Sep 15, 2026endpoint removed · crusoe12 → 11

Sep 11, 2026endpoint removed · nebius13 → 12

About

The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out). The Llama 3.3 instruction tuned text only model...

Specifications

Context window131,072

Max output16,384

Modalities intext

Modalities outtext

LicenseOpen weights

ReleasedDec 6, 2024

Knowledge cutoff2023-12-31

TokenizerLlama3

WeightsHugging Face ↗

Open weightsGATED

Hugging Facemeta-llama/Llama-3.3-70B-Instruct ↗

Downloads366K

Likes3.1K

Licensellama3.3

safetensorspytorch

Llama Guard 4 12B$0.18 / $0.18 Llama 4 Maverick$0.188 / $0.652 Llama 4 Scout$0.10 / $0.30 Llama 3.2 1B Instruct$0.027 / $0.201

Compare with

Llama 3.3 70B Instruct vs Mistral Large 4 Llama 3.3 70B Instruct vs Qwen3.8 Max Prime

Embed badge

Llama 3.3 70B Instruct price and speed badge

[![Llama 3.3 70B Instruct](https://modelindex.ai/badge/meta-llama/llama-3.3-70b-instruct)](https://modelindex.ai/models/meta-llama/llama-3.3-70b-instruct)

Sources

OpenRouter API23 minutes ago

AGGREGATOR

Hugging Face23 minutes ago

COMMUNITY

ModelIndex benchmarks3 days ago

MEASURED

Verified 23 minutes ago · list prices, not negotiated rates.