modelstatus.dev

Nemotron 3 Ultra

nvidia/nemotron-3-ultra-550b-a55bInputOutput

Compare its 4 providers Incident RSSCompare with Nemotron 3 Ultra (free), Nemotron 3 Nano 30B A3B, Nemotron 3 Super

CompareNemotron 3 Ultra

	State
BaseTenbaseten/fp4fp4	No data	—	—	99.97%	99.87%	—	—	—	203K	$0.600 / $2.40
DeepInfradeepinfra/fp4fp4	No data	—	—	99.91%	99.86%	—	—	—	262K	$0.500 / $2.20
Togethertogether	No data	—	—	—	—	—	—	—	512K	$0.600 / $3.60
Venicevenice/fp8fp8	No data	—	—	98.01%	97.83%	—	—	—	256K	$0.625 / $3.13

BaseTen—
No data24h —
7d
99.97%
30d
99.87%
p50
—
p99
—
Throughput
—
Context
203K
$/Mtok
$0.600 / $2.40
baseten/fp4fp4
DeepInfra—
No data24h —
7d
99.91%
30d
99.86%
p50
—
p99
—
Throughput
—
Context
262K
$/Mtok
$0.500 / $2.20
deepinfra/fp4fp4
Together—
No data24h —
7d
—
30d
—
p50
—
p99
—
Throughput
—
Context
512K
$/Mtok
$0.600 / $3.60
together
Venice—
No data24h —
7d
98.01%
30d
97.83%
p50
—
p99
—
Throughput
—
Context
256K
$/Mtok
$0.625 / $3.13
venice/fp8fp8

No recent measurement.

History

Availability

Axis starts at 90% so small differences stay visible; it expands to 0 when an endpoint falls below.

Time to first token

Logarithmic axis — the fleet spans two orders of magnitude. P50 is OpenRouter's rolling 30-minute window, averaged over the hour.

Throughput

Median tokens per second.

Daily availability

One row per endpoint · one dot per day · last 90 days.

Jul

Aug

Sep

Oct

worst hour ≥ 95%90–95%< 90%no data

Colour by each endpoint's worst hour of each UTC day · 4 endpoints · 47 days with data, 38 endpoint-days below 90%.