Name: Meta: Llama 3.3 70B Instruct API
Brand: Meta Llama
SKU: meta-llama/llama-3.3-70b-instruct

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).

#	Provider	Class	Input $/M	Output $/M	Cache ($/M Read)	vs consensus	TTFT	Throughput	Uptime 24h	Requests 24h
1	deepinfra	Direct host	$0.1	$0.32	—	-55.6% vs consensus	2.1s	15.6 tps	—	—
2	deepinfra → meta	Reseller	$0.1	$0.32	—	-55.6% vs consensus	—	—	—	—
3	openrouter → deepinfra	Reseller	$0.1	$0.32	—	-55.6% vs consensus	—	—	—	—
4	openrouter → meta	Reseller	$0.1	$0.32	—	-55.6% vs consensus	—	—	—	—
5	nebius	Direct host	$0.13	$0.4	—	-44.4% vs consensus	16.6s	4.6 tps	—	—
6	novita	Direct host	$0.14	$0.4	—	-44.4% vs consensus	1.7s	43.4 tps	—	—
7	novita → meta	Reseller	$0.14	$0.4	—	-44.4% vs consensus	—	—	—	—
8	openrouter → novita	Reseller	$0.14	$0.4	—	-44.4% vs consensus	—	—	—	—
9	parasail	Direct host	$0.22	$0.5	—	-30.6% vs consensus	2.7s	85.2 tps	—	—
10	openrouter → parasail	Reseller	$0.22	$0.5	$0.11	-30.6% vs consensus	—	—	—	—
11	openrouter → akashml	Reseller	$0.2	$0.52	$0.1	-27.8% vs consensus	—	—	—	—
12	coreweave	Direct host	$0.71	$0.71	—	-1.4% vs consensus	884ms	81.6 tps	—	—
13	azure → azure-ai	Reseller	$0.71	$0.71	—	-1.4% vs consensus	—	—	—	—
14	openrouter → coreweave	Reseller	$0.71	$0.71	$0.71	-1.4% vs consensus	—	—	—	—
15	wandb → meta	Reseller	$0.71	$0.71	—	-1.4% vs consensus	—	—	—	—
16	amazon-bedrock	Direct host	$0.72	$0.72	—	+0.0% vs consensus	—	—	—	—
17	bedrock	Direct host	$0.72	$0.72	—	+0.0% vs consensus	—	—	—	—
18	bedrock-converse	Direct host	$0.72	$0.72	—	+0.0% vs consensus	—	—	—	—
19	google	Direct host	$0.72	$0.72	—	+0.0% vs consensus	648ms	157.0 tps	—	—
20	oci	Direct host	$0.72	$0.72	—	+0.0% vs consensus	—	—	—	—
21	openrouter → google-vertex	Reseller	$0.72	$0.72	—	+0.0% vs consensus	—	—	—	—
22	ovhcloud	Direct host	$0.74	$0.74	—	+2.8% vs consensus	—	—	—	—
23	gmi	Direct host	$0.25	$0.75	—	+4.2% vs consensus	—	—	—	—
24	groq	Direct host	$0.59	$0.79	—	+9.7% vs consensus	825ms	296.3 tps	—	—
25	openrouter → groq	Reseller	$0.59	$0.79	$0.3	+9.7% vs consensus	—	—	—	—
26	fireworks	Direct host	$0.9	$0.9	—	+25.0% vs consensus	—	—	—	—
27	together	Direct host	$1.04	$1.04	—	+44.4% vs consensus	1.4s	76.3 tps	—	—
28	scaleway	Direct host	$1.05	$1.05	—	+45.8% vs consensus	1.4s	80.6 tps	—	—
29	sambanova	Direct host	$0.6	$1.2	—	+66.7% vs consensus	—	—	—	—
30	databricks	Direct host	$0.5	$1.5	—	+108.3% vs consensus	—	—	—	—
31	cloudflare	Direct host	$0.29	$2.25	—	+212.9% vs consensus	—	—	—	—
32	openrouter → cloudflare	Reseller	$0.29	$2.25	—	+212.9% vs consensus	—	—	—	—
33	compactifai Disabled by compactifai · stale	Direct host	$0.15	$0.31	—	-56.9% vs consensus	—	—	—	—
34	openrouter → nebius Disabled by openrouter · disabled · fp8	Reseller	$0.13	$0.4	—	-44.4% vs consensus	—	—	—	—
35	makora Disabled by makora · stale	Direct host	$0.18	$0.4	—	-44.4% vs consensus	—	—	—	—
36	hyperbolic Disabled by hyperbolic · stale	Direct host	$0.4	$0.4	—	-44.4% vs consensus	—	—	—	—
37	snowflake Disabled by snowflake · stale	Direct host	$0.58	$0.58	—	-19.4% vs consensus	—	—	—	—
38	friendli-ai Disabled by friendli-ai · stale	Direct host	$0.6	$0.6	—	-16.7% vs consensus	1.0s	135.6 tps	—	—
39	openrouter → wandb Disabled by openrouter · disabled · fp16	Reseller	$0.71	$0.71	$0.71	-1.4% vs consensus	—	—	—	—
40	openrouter → wandb-legacy Disabled by openrouter · disabled · fp16	Reseller	$0.71	$0.71	$0.71	-1.4% vs consensus	—	—	—	—
41	openrouter → crusoe Disabled by openrouter · disabled · bf16	Reseller	$0.25	$0.75	$0.13	+4.2% vs consensus	—	—	—	—
42	openrouter → sambanova-turbo Disabled by openrouter · disabled · bf16	Reseller	$0.45	$0.9	—	+25.0% vs consensus	—	—	—	—
43	openrouter → together Disabled by openrouter · disabled · fp8	Reseller	$1.04	$1.04	—	+44.4% vs consensus	—	—	—	—
44	openrouter → sambanova Disabled by openrouter · disabled · bf16	Reseller	$0.6	$1.2	—	+66.7% vs consensus	—	—	—	—

#
1
Provider
deepinfra
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.1
Output $ / 1M tokens
$0.32
Cache read $ / 1M tokens
—
-55.6% vs consensus
TTFT
2.1s
Throughput
15.6 tps
Uptime 24h
—
Requests 24h
—
#
2
Provider
deepinfra → meta
Class
Reseller
$ / 1M tokens
Input $ / 1M tokens
$0.1
Output $ / 1M tokens
$0.32
Cache read $ / 1M tokens
—
-55.6% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
3
Provider
openrouter → deepinfra
Class
Reseller
$ / 1M tokens
Input $ / 1M tokens
$0.1
Output $ / 1M tokens
$0.32
Cache read $ / 1M tokens
—
-55.6% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
4
Provider
openrouter → meta
Class
Reseller
$ / 1M tokens
Input $ / 1M tokens
$0.1
Output $ / 1M tokens
$0.32
Cache read $ / 1M tokens
—
-55.6% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
5
Provider
nebius
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.13
Output $ / 1M tokens
$0.4
Cache read $ / 1M tokens
—
-44.4% vs consensus
TTFT
16.6s
Throughput
4.6 tps
Uptime 24h
—
Requests 24h
—
#
6
Provider
novita
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.14
Output $ / 1M tokens
$0.4
Cache read $ / 1M tokens
—
-44.4% vs consensus
TTFT
1.7s
Throughput
43.4 tps
Uptime 24h
—
Requests 24h
—
#
7
Provider
novita → meta
Class
Reseller
$ / 1M tokens
Input $ / 1M tokens
$0.14
Output $ / 1M tokens
$0.4
Cache read $ / 1M tokens
—
-44.4% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
8
Provider
openrouter → novita
Class
Reseller
$ / 1M tokens
Input $ / 1M tokens
$0.14
Output $ / 1M tokens
$0.4
Cache read $ / 1M tokens
—
-44.4% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
9
Provider
parasail
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.22
Output $ / 1M tokens
$0.5
Cache read $ / 1M tokens
—
-30.6% vs consensus
TTFT
2.7s
Throughput
85.2 tps
Uptime 24h
—
Requests 24h
—
#
10
Provider
openrouter → parasail
Class
Reseller
$ / 1M tokens
Input $ / 1M tokens
$0.22
Output $ / 1M tokens
$0.5
Cache read $ / 1M tokens
$0.11
-30.6% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
11
Provider
openrouter → akashml
Class
Reseller
$ / 1M tokens
Input $ / 1M tokens
$0.2
Output $ / 1M tokens
$0.52
Cache read $ / 1M tokens
$0.1
-27.8% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
12
Provider
coreweave
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.71
Output $ / 1M tokens
$0.71
Cache read $ / 1M tokens
—
-1.4% vs consensus
TTFT
884ms
Throughput
81.6 tps
Uptime 24h
—
Requests 24h
—
#
13
Provider
azure → azure-ai
Class
Reseller
$ / 1M tokens
Input $ / 1M tokens
$0.71
Output $ / 1M tokens
$0.71
Cache read $ / 1M tokens
—
-1.4% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
14
Provider
openrouter → coreweave
Class
Reseller
$ / 1M tokens
Input $ / 1M tokens
$0.71
Output $ / 1M tokens
$0.71
Cache read $ / 1M tokens
$0.71
-1.4% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
15
Provider
wandb → meta
Class
Reseller
$ / 1M tokens
Input $ / 1M tokens
$0.71
Output $ / 1M tokens
$0.71
Cache read $ / 1M tokens
—
-1.4% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
16
Provider
amazon-bedrock
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.72
Output $ / 1M tokens
$0.72
Cache read $ / 1M tokens
—
+0.0% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
17
Provider
bedrock
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.72
Output $ / 1M tokens
$0.72
Cache read $ / 1M tokens
—
+0.0% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
18
Provider
bedrock-converse
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.72
Output $ / 1M tokens
$0.72
Cache read $ / 1M tokens
—
+0.0% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
19
Provider
google
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.72
Output $ / 1M tokens
$0.72
Cache read $ / 1M tokens
—
+0.0% vs consensus
TTFT
648ms
Throughput
157.0 tps
Uptime 24h
—
Requests 24h
—
#
20
Provider
oci
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.72
Output $ / 1M tokens
$0.72
Cache read $ / 1M tokens
—
+0.0% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
21
Provider
openrouter → google-vertex
Class
Reseller
$ / 1M tokens
Input $ / 1M tokens
$0.72
Output $ / 1M tokens
$0.72
Cache read $ / 1M tokens
—
+0.0% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
22
Provider
ovhcloud
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.74
Output $ / 1M tokens
$0.74
Cache read $ / 1M tokens
—
+2.8% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
23
Provider
gmi
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.25
Output $ / 1M tokens
$0.75
Cache read $ / 1M tokens
—
+4.2% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
24
Provider
groq
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.59
Output $ / 1M tokens
$0.79
Cache read $ / 1M tokens
—
+9.7% vs consensus
TTFT
825ms
Throughput
296.3 tps
Uptime 24h
—
Requests 24h
—
#
25
Provider
openrouter → groq
Class
Reseller
$ / 1M tokens
Input $ / 1M tokens
$0.59
Output $ / 1M tokens
$0.79
Cache read $ / 1M tokens
$0.3
+9.7% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
26
Provider
fireworks
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.9
Output $ / 1M tokens
$0.9
Cache read $ / 1M tokens
—
+25.0% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
27
Provider
together
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$1.04
Output $ / 1M tokens
$1.04
Cache read $ / 1M tokens
—
+44.4% vs consensus
TTFT
1.4s
Throughput
76.3 tps
Uptime 24h
—
Requests 24h
—
#
28
Provider
scaleway
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$1.05
Output $ / 1M tokens
$1.05
Cache read $ / 1M tokens
—
+45.8% vs consensus
TTFT
1.4s
Throughput
80.6 tps
Uptime 24h
—
Requests 24h
—
#
29
Provider
sambanova
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.6
Output $ / 1M tokens
$1.2
Cache read $ / 1M tokens
—
+66.7% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
30
Provider
databricks
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.5
Output $ / 1M tokens
$1.5
Cache read $ / 1M tokens
—
+108.3% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
31
Provider
cloudflare
Class
Direct host
$ / 1M tokens
Input $ / 1M tokens
$0.29
Output $ / 1M tokens
$2.25
Cache read $ / 1M tokens
—
+212.9% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
32
Provider
openrouter → cloudflare
Class
Reseller
$ / 1M tokens
Input $ / 1M tokens
$0.29
Output $ / 1M tokens
$2.25
Cache read $ / 1M tokens
—
+212.9% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
33
Provider
compactifai
Class
Direct host
Disabled by compactifai · stale
$ / 1M tokens
Input $ / 1M tokens
$0.15
Output $ / 1M tokens
$0.31
Cache read $ / 1M tokens
—
-56.9% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
34
Provider
openrouter → nebius
Class
Reseller
Disabled by openrouter · disabled · fp8
$ / 1M tokens
Input $ / 1M tokens
$0.13
Output $ / 1M tokens
$0.4
Cache read $ / 1M tokens
—
-44.4% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
35
Provider
makora
Class
Direct host
Disabled by makora · stale
$ / 1M tokens
Input $ / 1M tokens
$0.18
Output $ / 1M tokens
$0.4
Cache read $ / 1M tokens
—
-44.4% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
36
Provider
hyperbolic
Class
Direct host
Disabled by hyperbolic · stale
$ / 1M tokens
Input $ / 1M tokens
$0.4
Output $ / 1M tokens
$0.4
Cache read $ / 1M tokens
—
-44.4% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
37
Provider
snowflake
Class
Direct host
Disabled by snowflake · stale
$ / 1M tokens
Input $ / 1M tokens
$0.58
Output $ / 1M tokens
$0.58
Cache read $ / 1M tokens
—
-19.4% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
38
Provider
friendli-ai
Class
Direct host
Disabled by friendli-ai · stale
$ / 1M tokens
Input $ / 1M tokens
$0.6
Output $ / 1M tokens
$0.6
Cache read $ / 1M tokens
—
-16.7% vs consensus
TTFT
1.0s
Throughput
135.6 tps
Uptime 24h
—
Requests 24h
—
#
39
Provider
openrouter → wandb
Class
Reseller
Disabled by openrouter · disabled · fp16
$ / 1M tokens
Input $ / 1M tokens
$0.71
Output $ / 1M tokens
$0.71
Cache read $ / 1M tokens
$0.71
-1.4% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
40
Provider
openrouter → wandb-legacy
Class
Reseller
Disabled by openrouter · disabled · fp16
$ / 1M tokens
Input $ / 1M tokens
$0.71
Output $ / 1M tokens
$0.71
Cache read $ / 1M tokens
$0.71
-1.4% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
41
Provider
openrouter → crusoe
Class
Reseller
Disabled by openrouter · disabled · bf16
$ / 1M tokens
Input $ / 1M tokens
$0.25
Output $ / 1M tokens
$0.75
Cache read $ / 1M tokens
$0.13
+4.2% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
42
Provider
openrouter → sambanova-turbo
Class
Reseller
Disabled by openrouter · disabled · bf16
$ / 1M tokens
Input $ / 1M tokens
$0.45
Output $ / 1M tokens
$0.9
Cache read $ / 1M tokens
—
+25.0% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
43
Provider
openrouter → together
Class
Reseller
Disabled by openrouter · disabled · fp8
$ / 1M tokens
Input $ / 1M tokens
$1.04
Output $ / 1M tokens
$1.04
Cache read $ / 1M tokens
—
+44.4% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—
#
44
Provider
openrouter → sambanova
Class
Reseller
Disabled by openrouter · disabled · bf16
$ / 1M tokens
Input $ / 1M tokens
$0.6
Output $ / 1M tokens
$1.2
Cache read $ / 1M tokens
—
+66.7% vs consensus
TTFT
—
Throughput
—
Uptime 24h
—
Requests 24h
—

Source ecosystem

Loading…