Live leaderboard
Rankings that match your workload.
Public benchmarks (MMLU-Pro, GPQA, LiveBench, SWE-Bench) blended with our task-specific evals and real gateway traffic. Updated continuously.
Live·4,700 requests/sec across 161 tracked models
#
Model
Modality
01
claude-opus-4.5
Anthropic
textvision
94.2
82
320
$15.00↑
02
gpt-5.5
OpenAI
textvisionaudio
93.8
74
410
$12.50↑
03
claude-opus-4.1
Anthropic
textvision
93.1
78
340
$15.00·
04
gpt-5
OpenAI
textvisionaudio
92.4
78
420
$10.00·
05
gemini-3.1-pro
Google
textvisionaudiovideo
92.1
118
180
$7.00↑
06
gpt-5-codex
OpenAI
text
91.2
96
260
$4.50↑
07
claude-sonnet-4.5
Anthropic
textvision
90.8
118
220
$3.00↑
08
grok-5
xAI
textvision
90.6
95
260
$6.00·
09
gemini-2.5-pro
Google
textvisionaudiovideo
90.4
108
210
$5.00·
10
o3
OpenAI
textvision
90.1
62
520
$10.00·
11
deepseek-v4
DeepSeek
text
89.9
140
210
$0.55↑
12
claude-sonnet-4
Anthropic
textvision
89.4
122
210
$3.00·
13
qwen3-coder-480b
Alibaba
text
89.4
130
220
$1.20↑
14
grok-4
xAI
textvision
89.1
90
280
$5.00↓
15
qwen3-235b-thinking
Alibaba
text
88.9
82
340
$0.90↑
16
qwen3-max
Alibaba
textvision
88.4
128
240
$0.90↑
17
kimi-k2-thinking
Moonshot
text
88.4
82
360
$3.00↑
18
gpt-4.1
OpenAI
textvision
88.2
92
300
$8.00↓
19
deepseek-r1
DeepSeek
text
88.2
90
320
$0.55·
20
deepseek-coder-v3
DeepSeek
text
88.2
155
200
$0.60↑
21
claude-sonnet-3.7
Anthropic
textvision
88.1
128
200
$3.00↓
22
llama-4-405b
Meta
textvision
87.6
165
150
$2.40·
23
o4-mini
OpenAI
textvision
87.5
88
380
$1.10·
24
deepseek-v3.2-exp
DeepSeek
text
87.1
150
200
$0.28↑
25
qwen3-vl-235b
Alibaba
textvision
87
76
380
$2.00↑
26
llama-4-maverick
Meta
textvision
86.9
178
140
$1.80↑
27
nemotron-ultra-253b-v1
NVIDIA
text
86.7
120
260
$1.60↑
28
sonar-deep-research
Perplexity
text
86.5
60
640
$8.00↑
29
deepseek-v3.1-terminus
DeepSeek
text
86.4
148
205
$0.27·
30
gpt-4o
OpenAI
textvisionaudio
86.1
105
260
$5.00↓
31
qwen3-235b-a22b
Alibaba
text
86.1
145
210
$0.60↑
32
mistral-large-3
Mistral
text
86.1
110
280
$3.00↓
33
kimi-k2-0905
Moonshot
text
85.7
140
220
$2.50↑
34
deepseek-v3.1
DeepSeek
text
85.6
145
210
$0.27↓
35
gpt-5.5-mini
OpenAI
textvision
85.3
210
120
$0.40↑
36
llama-3.1-405b-instruct
Meta
text
85.2
130
200
$2.70↓
37
glm-4.6
Z.ai
text
85.1
160
200
$0.60↑
38
gemini-3.6-flash
Google
textvisionaudio
84.7
240
90
$0.30↑
39
minimax-m2
MiniMax
text
84.6
140
220
$1.20↑
40
o3-mini
OpenAI
text
84.4
92
340
$1.10↓
41
sonar-reasoning-pro
Perplexity
text
84.2
110
280
$8.00↑
42
gpt-5-mini
OpenAI
textvision
84.1
215
125
$0.35·
43
grok-3
xAI
text
84
92
300
$3.00↓
44
ernie-4.5-vl-424b
Baidu
textvision
84
100
300
$1.20↑
45
claude-haiku-4.5
Anthropic
textvision
83.9
195
130
$0.80↑
46
mistral-large-2411
Mistral
text
83.4
118
250
$2.00↓
47
glm-4.5
Z.ai
text
83.4
165
200
$0.50·
48
doubao-seed-1.6
ByteDance
textvision
83.2
160
210
$0.80↑
49
deepseek-coder-v2
DeepSeek
text
83.1
170
180
$0.40↓
50
gemini-2.5-flash
Google
textvisionaudio
82.9
260
85
$0.30·
51
command-a-03-2025
Cohere
text
82.9
130
220
$2.50↑
52
gpt-4.1-mini
OpenAI
textvision
82.6
200
130
$0.40↓
53
qwen3-coder-30b
Alibaba
text
82.6
220
110
$0.30↑
54
nova-pro-v1
Amazon
textvisionvideo
82.6
140
210
$3.20·
55
deepseek-r1-distill-70b
DeepSeek
text
82.5
165
180
$0.35·
56
qwen3-vl-72b
Alibaba
textvision
82.4
88
340
$1.20·
57
ernie-4.5-300b
Baidu
text
82.4
130
240
$0.70·
58
hermes-4-405b
Nous
text
82.4
130
230
$1.80↑
59
llama-4-scout
Meta
textvision
82.1
240
100
$0.60↑
60
codestral-2508
Mistral
text
82.1
200
120
$0.40↑
61
step-3
StepFun
textvision
82.1
130
240
$1.00↑
62
grok-code-fast-1
xAI
text
82
260
90
$0.60↑
63
magistral-medium-1.2
Mistral
text
82
120
260
$1.50↑
64
qwq-32b
Alibaba
text
81.6
130
240
$0.30↓
65
sonar-pro
Perplexity
text
81.6
130
240
$3.00↑
66
mistral-medium-3.1
Mistral
textvision
81.2
140
200
$0.90·
67
llama-4-70b
Meta
text
81.1
220
110
$0.90·
68
glm-4.5v
Z.ai
textvision
80.6
130
240
$0.90↑
69
llama-3.3-70b-instruct
Meta
text
80.4
225
115
$0.70↓
70
devstral-medium
Mistral
text
80.4
190
130
$0.50↑
71
nemotron-super-49b-v1.5
NVIDIA
text
80.4
180
150
$0.50↑
72
gemini-2.0-flash
Google
textvisionaudio
80.3
250
95
$0.30↓
73
qwen2.5-coder-32b
Alibaba
text
80.2
210
120
$0.30↓
74
tulu-3-405b
AllenAI
text
80.2
130
240
$2.40·
75
hunyuan-large
Tencent
text
80.1
140
220
$1.20·
76
grok-5-fast
xAI
text
79.8
280
80
$0.50↑
77
minimax-m1
MiniMax
text
79.8
150
210
$0.60·
78
claude-haiku-3.5
Anthropic
textvision
79.6
220
110
$0.80·
79
pixtral-large
Mistral
textvision
79.6
100
320
$2.00·
80
reka-core
Reka
textvisionvideo
79.6
130
240
$2.00·
81
llama-3.2-90b-vision
Meta
textvision
79.4
150
200
$0.90·
82
mai-1-preview
Microsoft
text
79.4
150
220
$2.50↑
83
internvl3-78b
OpenGVLab
textvision
79.4
100
300
$0.80↑
84
qwen3-omni-30b
Alibaba
textvisionaudio
79.2
180
150
$0.40↑
85
command-r-plus-08-2024
Cohere
text
79.1
150
200
$2.50↓
86
qwen2.5-72b-instruct
Alibaba
text
78.9
190
140
$0.35↓
87
yi-large
01.AI
text
78.9
150
210
$3.00↓
88
longcat-flash-thinking
Meituan
text
78.9
150
220
$0.50↑
89
gemini-3-flash-lite
Google
textvision
78.8
320
70
$0.10↑
90
hermes-4-70b
Nous
text
78.6
200
130
$0.50↑
91
llama-3.1-70b-instruct
Meta
text
78.5
240
100
$0.60↓
92
gpt-5.5-nano
OpenAI
text
78.4
320
70
$0.10↑
93
qwen3-32b
Alibaba
text
78.4
200
130
$0.20·
94
jamba-1.7-large
AI21
text
78.4
140
210
$2.00·
95
gpt-4o-mini
OpenAI
textvision
78.2
240
110
$0.15↓
96
grok-4-fast
xAI
text
77.8
300
75
$0.40·
97
glm-4.5-air
Z.ai
text
77.1
250
100
$0.20·
98
gpt-4.1-nano
OpenAI
text
76.9
300
80
$0.10↓
99
kimi-linear-48b
Moonshot
text
76.8
240
100
$0.30↑
100
mistral-small-3.2
Mistral
textvision
76.5
220
110
$0.20·
101
nova-lite-v1
Amazon
textvisionvideo
76.4
260
95
$0.24·
102
seed-oss-36b-instruct
ByteDance
text
76.4
200
130
$0.30↑
103
solar-pro-2
Upstage
text
76.4
200
130
$0.50↑
104
aria-25b
Rhymes
textvision
76.4
140
220
$0.50·
105
phi-4
Microsoft
text
76.2
240
110
$0.15·
106
gemini-2.5-flash-lite
Google
text
76.1
340
65
$0.10·
107
deepcoder-14b-preview
AgenticaAI
text
76.1
240
100
$0.20↑
108
magistral-small-1.2
Mistral
text
75.4
200
130
$0.30↑
109
molmo-72b
AllenAI
textvision
75.1
100
300
$0.90·
110
qwen3-14b
Alibaba
text
74.6
260
100
$0.15·
111
aya-expanse-32b
Cohere
text
74.6
200
130
$0.30·
112
inflection-3-productivity
Inflection
text
74.6
180
160
$2.50↓
113
wizardcoder-33b
WizardLM
text
74.6
200
130
$0.30↓
114
gemma-3-27b-it
Google
textvision
74.5
180
140
$0.20·
115
reka-flash-3
Reka
textvision
74.4
220
110
$0.40·
116
lfm-40b
LiquidAI
text
74.4
240
100
$0.30·
117
ernie-4.5-21b-a3b
Baidu
text
74.2
260
100
$0.20·
118
deephermes-3-mistral-24b
Nous
text
74.2
220
110
$0.30·
119
dbrx-instruct
Databricks
text
74.2
180
150
$1.50↓
120
longcat-flash-chat
Meituan
text
74.2
220
110
$0.30↑
121
phi-4-multimodal
Microsoft
textvisionaudio
74.1
200
130
$0.20·
122
sonar
Perplexity
text
74.1
200
140
$1.00·
123
goliath-120b
Alpindale
text
74.1
130
260
$1.50↓
124
llava-onevision-72b
LMMs-Lab
textvision
74.1
100
300
$0.80·
125
grok-3-mini
xAI
text
74
260
90
$0.30↓
126
hunyuan-a13b-instruct
Tencent
text
73.8
250
100
$0.20·
127
pixtral-12b
Mistral
textvision
72.8
180
160
$0.20↓
128
ui-tars-1.5-7b
ByteDance
textvision
72.6
240
110
$0.20↑
129
yi-lightning
01.AI
text
72.6
280
90
$0.14·
130
nous-capybara-34b
Nous
text
72.6
220
120
$0.40↓
131
olmo-2-32b-instruct
AllenAI
text
72.4
200
130
$0.20·
132
rakuten-ai-2.0-8x7b
Rakuten
text
72.4
200
130
$0.40·
133
starcoder2-15b-instruct
BigCode
text
72.4
240
100
$0.20·
134
yi-1.5-200k
01.AI
text
72.4
130
260
$0.60↓
135
llama-3.2-11b-vision
Meta
textvision
72.1
260
100
$0.20·
136
jamba-1.7-mini
AI21
text
72.1
240
110
$0.20·
137
lzlv-70b-fp16-hf
lizpreciatior
text
71.8
190
140
$0.60↓
138
nemotron-nano-9b-v2
NVIDIA
text
71.4
320
80
$0.10↑
139
step-2-mini
StepFun
text
71.4
280
90
$0.15·
140
inflection-3-pi
Inflection
text
71.2
200
140
$2.50↓
141
granite-3.2-vision
IBM
textvision
71.2
220
120
$0.20·
142
internlm3-8b-instruct
InternLM
text
71.2
320
80
$0.10·
143
phi-4-mini
Microsoft
text
70.8
340
75
$0.08·
144
sarashina2-vision-14b
Sakana
textvision
70.8
220
120
$0.30·
145
arctic-instruct
Snowflake
text
70.6
200
140
$1.00↓
146
command-r7b-12-2024
Cohere
text
70.4
320
80
$0.15·
147
airoboros-l2-70b
Jon Durbin
text
70.4
200
130
$0.50↓
148
gemma-3-12b-it
Google
text
70.2
240
100
$0.10·
149
falcon3-10b-instruct
TII
text
70.2
280
90
$0.15·
150
qwen3-8b
Alibaba
text
70.1
320
80
$0.10·
151
nova-micro-v1
Amazon
text
70.1
340
70
$0.14·
152
ministral-8b
Mistral
text
68.5
340
65
$0.10↓
153
granite-3.3-8b-instruct
IBM
text
68.4
340
70
$0.10·
154
openchat-3.6-8b
OpenChat
text
68.4
340
70
$0.05↓
155
llama-3.1-8b-instruct
Meta
text
68.2
380
55
$0.05↓
156
lfm-7b
LiquidAI
text
68.2
380
60
$0.05·
157
falcon-mamba-7b
TII
text
66.8
340
70
$0.10·
158
mythomax-l2-13b
Gryphe
text
66.2
300
80
$0.10↓
159
llama-guard-3-8b
Meta
text
65
340
60
$0.05·
160
toppy-m-7b
Undi95
text
62.4
380
60
$0.07↓
161
ministral-3b
Mistral
text
62.1
420
45
$0.04↓
Preview data · Quality = weighted average of MMLU-Pro, GPQA-Diamond, LiveBench, SWE-Bench-Verified, and our proprietary router-eval.
Top apps & agents on llmcloud.ai
Ranked by tokens routed through the gateway over the last 7 days.
| # | App | Category | Share of tokens | Tokens / 7d | Routes to |
|---|---|---|---|---|---|
| 1 | Cline Autonomous coding agent inside VS Code. | Coding agents | 412B | auto:coding | |
| 2 | Roo Code Multi-mode coding agent with architect and debug personas. | Coding agents | 268B | auto:coding | |
| 3 | SillyTavern Power-user front end for persona and story chat. | Roleplay & companions | 231B | auto:chat | |
| 4 | Janitor AI Hosted character chat with bring-your-own-key. | Roleplay & companions | 187B | auto:chat | |
| 5 | Kilo Code Open-source agent merging Cline and Roo workflows. | Coding agents | 141B | auto:coding | |
| 6 | LiteLLM Proxy and SDK that normalizes 100+ providers. | Infra & proxies | 118B | auto | |
| 7 | Aider Terminal pair programmer that commits as it goes. | Coding agents | 96B | auto:coding | |
| 8 | Continue Open-source autopilot for VS Code and JetBrains. | IDEs & editors | 88B | auto:coding | |
| 9 | Open WebUI Self-hosted chat UI for teams. | Productivity | 79B | auto | |
| 10 | OpenHands Sandboxed agent that runs a real shell and browser. | Autonomous agents | 74B | auto:agent |