Meta: Llama 3.1 70B Instruct

meta-llama
freeoutput / 1M tokens

Best price — Output ($/1M tokens)

Markets

One row per access path (venue → underlying provider). Prices are each venue's published list price; performance is measured by Artificial Analysis. Your actual bill may vary by routing settings, fallbacks, region / ZDR, long-context tiers, and request features (JSON mode, tools, vision).

#ProviderClassInput $/MOutput $/MCache ($/M Read)vs consensusTTFTThroughputUptime 24hRequests 24h
1
google-vertex
Direct hostfree
free
—-100.0% vs consensus
—
———
2
deepinfra
Direct host$0.4
$0.4
—-42.4% vs consensus
2.0s
37.4 tps——
3
deepinfra → meta
Reseller$0.4
$0.4
—-42.4% vs consensus
—
———
4
openrouter → deepinfra
Reseller$0.4
$0.4
—-42.4% vs consensus
—
———
5
openrouter → meta
Reseller$0.4
$0.4
—-42.4% vs consensus
—
———
6
ovhcloud
Direct host$0.67
$0.67
—-3.6% vs consensus
—
———
7
oci
Direct host$0.72
$0.72
—+3.6% vs consensus
—
———
8
openrouter → amazon-bedrock
Reseller$0.72
$0.72
—+3.6% vs consensus
—
———
9
openrouter → coreweave
Reseller$0.8
$0.8
$0.8+15.1% vs consensus
—
———
10
amazon-bedrock
Direct host$0.9
$0.9
—+29.5% vs consensus
1.3s
128.5 tps——
11
bedrock
Direct host$0.99
$0.99
—+42.4% vs consensus
—
———
12
azure → azure-ai
Reseller$2.68
$3.54
—+409.4% vs consensus
—
———
13
openrouter → wandb
Disabled by openrouter · disabled · bf16
Reseller$0.8
$0.8
$0.8+15.1% vs consensus
—
———
14
openrouter → wandb-legacy
Disabled by openrouter · disabled · bf16
Reseller$0.8
$0.8
$0.8+15.1% vs consensus
—
———
Source ecosystem
Loading…