Meta

42 models from Meta, ranked by Podium Score.

Creator of Llama open-weight models. Meta AI releases powerful open-source language models.

Muse Spark 1.1
Top model
69.9
Podium Score
39.1
Average score
17
open weights

Meta API Pricing — per 1M Tokens

Get API Key →
ModelIn $/MOut $/MSpeedContextPodium ScoreBenchmarks
Muse Spark 1.1
$1.25/M$4.25/M208 t/s1.0M69.9Benchmarks →
Llama 3.1 Nemotron Ultra 253b V1
$0/M$0/M50 t/s128K67.0Benchmarks →
Llama 3.1 405b Instruct Bf16
$0/M$0/M50 t/s128K67.0Benchmarks →
Llama 3.1 405b Instruct Fp8
$0/M$0/M50 t/s128K67.0Benchmarks →
Llama 3.3 Nemotron 49b Super V1
$0/M$0/M50 t/s128K66.0Benchmarks →
Llama 4 Maverick 17b 128e Instruct
$0/M$0/M50 t/s128K66.0Benchmarks →
Llama 4 Scout 17b 16e Instruct
$0/M$0/M50 t/s128K66.0Benchmarks →
Llama 3.3 70b Instruct
$0/M$0/M50 t/s128K66.0Benchmarks →
Llama 3.1 Nemotron 70b Instruct
$0/M$0/M50 t/s128K65.0Benchmarks →
Llama 3.1 70b Instruct
$0/M$0/M50 t/s128K65.0Benchmarks →
Llama 3.1 Nemotron 51b Instruct
$0/M$0/M50 t/s128K64.0Benchmarks →
Llama 3.1 Tulu 3 70b
$0/M$0/M50 t/s128K64.0Benchmarks →
Llama 3 70b Instruct
$0/M$0/M50 t/s128K64.0Benchmarks →
Llama 3 8b Instruct
$0/M$0/M50 t/s128K61.0Benchmarks →
Llama 3.1 Tulu 3 8b
$0/M$0/M50 t/s128K61.0Benchmarks →
Llama 3.1 8b Instruct
$0/M$0/M50 t/s128K61.0Benchmarks →
Llama 2 70b Chat
$0/M$0/M50 t/s128K59.0Benchmarks →
Llama 3.2 3b Instruct
$0/M$0/M50 t/s128K58.0Benchmarks →
Llama2 70b Steerlm Chat
$0/M$0/M50 t/s128K58.0Benchmarks →
Llama 2 13b Chat
$0/M$0/M50 t/s128K57.0Benchmarks →
Muse Spark 1 2
$0/M$0/M80 t/s131K56.8Benchmarks →
Llama 3.2 1b Instruct
$0/M$0/M50 t/s128K56.0Benchmarks →
Llama 2 7b Chat
$0/M$0/M50 t/s128K55.0Benchmarks →
Muse Spark
$0/M$0/M80 t/s1.0M47.0Benchmarks →
Muse Glimmer
$0/M$0/M80 t/s131K35.1Benchmarks →
Llama Nemotron Super 49b V1 5 Reasoning
$0/M$0/M50 t/s131K12.4Benchmarks →
Llama 4 Scout
$0/M$0/M80 t/s131K10.3Benchmarks →
Hermes 4 Llama 3 1 70b Reasoning
$0/M$0/M50 t/s131K9.9Benchmarks →
Llama 3 3 Instruct 70b
$0/M$0/M80 t/s131K9.3Benchmarks →
Llama 3 1 Nemotron Ultra 253b V1 Reasoning
$0/M$0/M50 t/s131K8.9Benchmarks →
Hermes 4 Llama 3 1 405b Reasoning
$0/M$0/M50 t/s131K8.8Benchmarks →
Hermes 4 Llama 3 1 405b
$0/M$0/M80 t/s131K8.6Benchmarks →
Llama Nemotron Super 49b V1 5
$0/M$0/M80 t/s131K8.5Benchmarks →
Llama 3 1 Instruct 8b
$0/M$0/M80 t/s131K7.4Benchmarks →
Llama 3 1 Nemotron Instruct 70b
$0/M$0/M80 t/s131K7.4Benchmarks →
Hermes 4 Llama 3 1 70b
$0/M$0/M80 t/s131K6.7Benchmarks →
Llama 3 1 Instruct 70b
$0/M$0/M80 t/s131K6.5Benchmarks →
Hermes 3 Llama 3 1 70b
$0/M$0/M80 t/s131K4.8Benchmarks →
Llama 2 Chat 7b
$0/M$0/M80 t/s131K3.9Benchmarks →
Llama 3 Instruct 70b
$0/M$0/M80 t/s131K3.1Benchmarks →
Llama 3 2 Instruct 11b Vision
$0/M$0/M80 t/s131K3.0Benchmarks →
Llama 3 Instruct 8b
$0/M$0/M80 t/s131K1.0Benchmarks →
Pricing and speed from official API documentation; scores aggregated from five leaderboards.
Muse Spark 1.1
69.9
1.0M ctx208 t/s$4.25/M out
69.9
Llama 3.1 Nemotron Ultra 253b V1
67.0
128K ctx50 t/s$0/M out
67.0
Llama 3.1 405b Instruct Bf16
67.0
128K ctx50 t/s$0/M out
67.0
Llama 3.1 405b Instruct Fp8
67.0
128K ctx50 t/s$0/M out
67.0
Llama 3.3 Nemotron 49b Super V1
66.0
128K ctx50 t/s$0/M out
66.0
Llama 4 Maverick 17b 128e Instruct
66.0
128K ctx50 t/s$0/M out
66.0
Llama 4 Scout 17b 16e Instruct
66.0
128K ctx50 t/s$0/M out
66.0
Llama 3.3 70b Instruct
66.0
128K ctx50 t/s$0/M out
66.0
Llama 3.1 Nemotron 70b Instruct
65.0
128K ctx50 t/s$0/M out
65.0
Llama 3.1 70b Instruct
65.0
128K ctx50 t/s$0/M out
65.0
Llama 3.1 Nemotron 51b Instruct
64.0
128K ctx50 t/s$0/M out
64.0
Llama 3.1 Tulu 3 70b
64.0
128K ctx50 t/s$0/M out
64.0
Llama 3 70b Instruct
64.0
128K ctx50 t/s$0/M out
64.0
Llama 3 8b Instruct
61.0
128K ctx50 t/s$0/M out
61.0
Llama 3.1 Tulu 3 8b
61.0
128K ctx50 t/s$0/M out
61.0
Llama 3.1 8b Instruct
61.0
128K ctx50 t/s$0/M out
61.0
Llama 2 70b Chat
59.0
128K ctx50 t/s$0/M out
59.0
Llama 3.2 3b Instruct
58.0
128K ctx50 t/s$0/M out
58.0
Llama2 70b Steerlm Chat
58.0
128K ctx50 t/s$0/M out
58.0
Llama 2 13b Chat
57.0
128K ctx50 t/s$0/M out
57.0
Muse Spark 1 2
56.8
131K ctx80 t/s$0/M out
56.8
Llama 3.2 1b Instruct
56.0
128K ctx50 t/s$0/M out
56.0
Llama 2 7b Chat
55.0
128K ctx50 t/s$0/M out
55.0
Muse Spark
47.0
1.0M ctx80 t/s$0/M out
47.0
Muse Glimmer
35.1
131K ctx80 t/s$0/M out
35.1
Llama Nemotron Super 49b V1 5 Reasoning
12.4
131K ctx50 t/s$0/M out
12.4
Open
Llama 4 Scout
10.3
131K ctx80 t/s$0/M out
10.3
Open
Hermes 4 Llama 3 1 70b Reasoning
9.9
131K ctx50 t/s$0/M out
9.9
Open
Llama 3 3 Instruct 70b
9.3
131K ctx80 t/s$0/M out
9.3
Open
Llama 3 1 Nemotron Ultra 253b V1 Reasoning
8.9
131K ctx50 t/s$0/M out
8.9
Open
Hermes 4 Llama 3 1 405b Reasoning
8.8
131K ctx50 t/s$0/M out
8.8
Open
Hermes 4 Llama 3 1 405b
8.6
131K ctx80 t/s$0/M out
8.6
Open
Llama Nemotron Super 49b V1 5
8.5
131K ctx80 t/s$0/M out
8.5
Open
Llama 3 1 Instruct 8b
7.4
131K ctx80 t/s$0/M out
7.4
Open
Llama 3 1 Nemotron Instruct 70b
7.4
131K ctx80 t/s$0/M out
7.4
Open
Hermes 4 Llama 3 1 70b
6.7
131K ctx80 t/s$0/M out
6.7
Open
Llama 3 1 Instruct 70b
6.5
131K ctx80 t/s$0/M out
6.5
Open
Hermes 3 Llama 3 1 70b
4.8
131K ctx80 t/s$0/M out
4.8
Open
Llama 2 Chat 7b
3.9
131K ctx80 t/s$0/M out
3.9
Open
Llama 3 Instruct 70b
3.1
131K ctx80 t/s$0/M out
3.1
Open
Llama 3 2 Instruct 11b Vision
3.0
131K ctx80 t/s$0/M out
3.0
Open
Llama 3 Instruct 8b
1.0
131K ctx80 t/s$0/M out
1.0
Open
0 / 4
Compare Now →