Meta: Llama 3.3 70B Instruct

meta-llama/llama-3.3-70b-instruct

Llama 3.3 70B Instruct is a 70B open-weight model, ranking #166 of 234 on the TryAii Score. Its best showing is #2 on GSM8K. It's served by 14 independent hosts from $0.1/$0.32 per million input/output tokens — about a third of what models at its level usually charge.

Prompt $/1M

$0.100

Completion $/1M

$0.320

Context

131,072 tok

Best provider

176 tok/s · Groq

Specs

Model id: meta-llama/llama-3.3-70b-instruct
Display name: Meta: Llama 3.3 70B Instruct
Modality: text->text
Tokenizer: Llama3
Instruct type: llama3
Context length: 131,072
Max completion: 16,384
Prompt $/1M: $0.100
Completion $/1M: $0.320
Free tier: no

Capabilities

Function callingno
JSON modeno
System promptyes
Visionno
Moderatedno

Benchmark scores

Each axis is one benchmark, normalized to the best model in the field.

No benchmark data for this model.

Speed by provider

Provider	tok/s	TTFT	Total latency
Groqfastest	176	206 ms	274 ms
Artificial Analysis	86	608 ms	—
Friendli	62	182 ms	188 ms
CoreWeave	51	321 ms	337 ms
WandB	40	181 ms	217 ms
Crusoe	38	446 ms	522 ms
Parasail	32	681 ms	876 ms
SambaNova	30	720 ms	826 ms
Cloudflare	28	508 ms	633 ms
Google	28	421 ms	444 ms
Novita	25	646 ms	708 ms
Together	18	1,091 ms	1,396 ms
AkashML	18	957 ms	1,206 ms
DeepInfra	15	462 ms	655 ms
Nebius	9	714 ms	2,159 ms
Inceptron	7	645 ms	857 ms