Fleet Operations

Discover

Live operational intelligence across every free NVIDIA NIM endpoint — ranked by throughput, reliability, and congestion, refreshed continuously.

46
Endpoints
8
Providers
21
Degraded
just now
Last probe
Fleet Degraded25 of 46 endpoints operational · 21 degraded
Throughput▼6.0%
35.2tok/s
P95 Latency▲34%
3,316ms
Reliability▼2.5pt
88.0%
Congestion▲2.5pt
23%
Model Registry
46 endpoints
Fleet Performance · 12h
836ms
▲ 2.7% · 12h
avg930min395max2230

Loading…

Provider Intelligence
8 providers · 46 models · 21 degraded
ProviderMCongestionUptimeDegRouteRecovery
AAlibaba1
6%
100.00%
0
0/1
CGoogle4
32%
79.17%
1
0/4
DOther10
26%
89.83%
6
0/10
DNVIDIA15
17%
91.11%
6
3/15
DMistral4
9%
94.17%
2
0/4
DMicrosoft1
100%
35.00%
1
0/1
DMeta9
18%
94.26%
3
0/9
DDeepSeek2
61%
55.00%
2
0/2
Overall Rankings
46 ranked · throughput · reliability · congestion
Runner Up
nemotron-3-nano-30b-a3b
NVIDIA
628 pts
Best Overall
mistral-small-4-119b-2603
Mistral
746 pts
Third
step-3.7-flash
Other
627 pts
1
mistral-small-4-119b-2603
Mistral
746A
2
nemotron-3-nano-30b-a3b
NVIDIA
628A
3
step-3.7-flash
Other
627A
4
nemotron-3-nano-omni-30b-a3b-reasoning
NVIDIA
596A
5
nemotron-mini-4b-instruct
NVIDIA
595A
6
gpt-oss-20b
Other
589A
7
diffusiongemma-26b-a4b-it
Google
579A
8
llama-4-maverick-17b-128e-instruct
Meta
544A
9
nvidia-nemotron-nano-9b-v2
NVIDIA
538A
10
nemotron-3-content-safety
NVIDIA
529A
11
riva-translate-4b-instruct-v1.1
NVIDIA
525A
12
qwen3-next-80b-a3b-instruct
Alibaba
519A
13
llama-3.1-nemotron-safety-guard-8b-v3
NVIDIA
501A
14
sarvam-m
Other
497A
15
kimi-k2.6
Other
494A
16
llama-3.1-8b-instruct
Meta
493A
17
llama-3.3-nemotron-super-49b-v1
NVIDIA
492A
18
solar-10.7b-instruct
Other
489A
19
llama-3.2-3b-instruct
Meta
489A
20
nemotron-3.5-content-safety
NVIDIA
488A
21
stockmark-2-100b-instruct
Other
480A
22
ministral-14b-instruct-2512
Mistral
479A
23
gpt-oss-120b
Other
477A
24
llama-3.2-1b-instruct
Meta
475A
25
llama-3.1-nemoguard-8b-content-safety
NVIDIA
470A
26
gemma-3n-e4b-it
Google
470A
27
llama-3.2-90b-vision-instruct
Meta
468A
28
gemma-3n-e2b-it
Google
465A
29
llama-3.2-11b-vision-instruct
Meta
464A
30
step-3.5-flash
Other
457A
31
llama-3.1-nemotron-nano-vl-8b-v1
NVIDIA
455A
32
llama-guard-4-12b
Meta
449A
33
mistral-large-3-675b-instruct-2512
Mistral
445A
34
mistral-medium-3.5-128b
Mistral
434A
35
nemotron-3-super-120b-a12b
NVIDIA
432A
36
llama-3.1-70b-instruct
Meta
394A
37
llama-3.3-nemotron-super-49b-v1.5
NVIDIA
366A
38
deepseek-v4-pro
DeepSeek
315A
39
llama-3.3-70b-instruct
Meta
249C
40
llama-3.1-nemotron-nano-8b-v1
NVIDIA
224C
41
glm-5.1
Other
174D
42
nemotron-3-ultra-550b-a55b
NVIDIA
174D
43
deepseek-v4-flash
DeepSeek
144D
44
minimax-m2.7
Other
120D
45
phi-4-mini-instruct
Microsoft
105D
46
gemma-4-31b-it
Google
51D
Trend Analysis
46 models tracked
Rising2
nemotron-mini-4b-instruct100%
nemotron-3-nano-omni-30b-a3b-reasoning100%
Improving17
glm-5.158%
nemotron-3-super-120b-a12b95%
llama-3.3-nemotron-super-49b-v1.573%
llama-3.3-nemotron-super-49b-v197%
llama-3.1-nemotron-nano-8b-v143%
Stable27
solar-10.7b-instruct100%
stockmark-2-100b-instruct100%
step-3.7-flash100%
step-3.5-flash100%
sarvam-m100%
Declining0
All clear
Reliability Matrix
by routing confidence
solar-10.7b-instruct
100%High
stockmark-2-100b-instruct
100%Moderate
step-3.7-flash
100%Moderate
step-3.5-flash
100%Moderate
sarvam-m
100%High
qwen3-next-80b-a3b-instruct
100%Moderate
gpt-oss-20b
100%High
gpt-oss-120b
100%Moderate
riva-translate-4b-instruct-v1.1
100%Moderate
nvidia-nemotron-nano-9b-v2
100%Moderate
nemotron-mini-4b-instruct
100%Moderate
nemotron-3.5-content-safety
100%Moderate
nemotron-3-nano-omni-30b-a3b-reasoning
100%Moderate
nemotron-3-nano-30b-a3b
100%High
nemotron-3-content-safety
100%Moderate
llama-3.1-nemotron-safety-guard-8b-v3
100%High
llama-3.1-nemotron-nano-vl-8b-v1
100%Moderate
llama-3.1-nemoguard-8b-content-safety
100%Moderate
kimi-k2.6
100%Moderate
mistral-small-4-119b-2603
100%High
llama-4-maverick-17b-128e-instruct
100%Moderate
gemma-3n-e4b-it
100%Moderate
gemma-3n-e2b-it
100%Moderate
diffusiongemma-26b-a4b-it
100%High
ministral-14b-instruct-2512
98%Moderate
llama-guard-4-12b
98%Moderate
llama-3.2-90b-vision-instruct
98%Moderate
llama-3.2-3b-instruct
98%Moderate
llama-3.2-11b-vision-instruct
98%Moderate
llama-3.1-8b-instruct
98%Moderate
llama-3.3-nemotron-super-49b-v1
97%Moderate
nemotron-3-super-120b-a12b
95%Moderate
mistral-medium-3.5-128b
90%Moderate
mistral-large-3-675b-instruct-2512
88%Moderate
llama-3.2-1b-instruct
88%Moderate
llama-3.1-70b-instruct
85%Moderate
llama-3.3-70b-instruct
83%Avoid
llama-3.3-nemotron-super-49b-v1.5
73%Avoid
deepseek-v4-pro
62%Avoid
glm-5.1
58%Avoid
nemotron-3-ultra-550b-a55b
58%Avoid
deepseek-v4-flash
48%Avoid
llama-3.1-nemotron-nano-8b-v1
43%Avoid
minimax-m2.7
40%Avoid
phi-4-mini-instruct
35%Avoid
gemma-4-31b-it
17%Avoid
Incident Timeline
21 active
Congestion spike detected — consider failover
glm-5.1 · Other
1m ago
Congestion spike detected — consider failover
nemotron-3-ultra-550b-a55b · NVIDIA
< 1m ago
Congestion spike detected — consider failover
llama-3.3-nemotron-super-49b-v1.5 · NVIDIA
< 1m ago
Congestion spike detected — consider failover
llama-3.1-nemotron-nano-8b-v1 · NVIDIA
< 1m ago
Congestion spike detected — consider failover
minimax-m2.7 · Other
< 1m ago
Congestion spike detected — consider failover
phi-4-mini-instruct · Microsoft
< 1m ago
Congestion spike detected — consider failover
llama-3.3-70b-instruct · Meta
< 1m ago
Congestion spike detected — consider failover
gemma-4-31b-it · Google
< 1m ago
Congestion spike detected — consider failover
deepseek-v4-pro · DeepSeek
< 1m ago
Congestion spike detected — consider failover
deepseek-v4-flash · DeepSeek
< 1m ago
Throughput degradation on primary endpoint
step-3.7-flash · Other
1m ago
Throughput degradation on primary endpoint
step-3.5-flash · Other
1m ago
Throughput degradation on primary endpoint
gpt-oss-120b · Other
1m ago
Throughput degradation on primary endpoint
nvidia-nemotron-nano-9b-v2 · NVIDIA
1m ago
Throughput degradation on primary endpoint
nemotron-3-super-120b-a12b · NVIDIA
1m ago
Throughput degradation on primary endpoint
llama-3.1-nemotron-nano-vl-8b-v1 · NVIDIA
< 1m ago
Throughput degradation on primary endpoint
kimi-k2.6 · Other
< 1m ago
Throughput degradation on primary endpoint
mistral-medium-3.5-128b · Mistral
< 1m ago
Throughput degradation on primary endpoint
mistral-large-3-675b-instruct-2512 · Mistral
< 1m ago
Throughput degradation on primary endpoint
llama-3.2-1b-instruct · Meta
< 1m ago
Throughput degradation on primary endpoint
llama-3.1-70b-instruct · Meta
< 1m ago