modelstatus.dev

Inkling Small

thinkingmachines/inkling-small

Compare with Inkling, Inkling (batch), GLM 5.2

CompareInkling Small

	State
DeepInfradeepinfra/fp8fp8	Healthy	100%	98.07%	98.30%	98.30%	414ms	1.52s	165 tps	524K	$0.450 / $1.20
BaseTenbaseten/fp8fp8	No data	—	98.94%	97.68%	97.68%	377ms	1.99s	202 tps	1.0M	$0.500 / $1.20
Togethertogether	No data	—	98.85%	94.51%	94.51%	446ms	6.20s	89 tps	524K	$0.500 / $1.20

DeepInfra100%
Healthy24h 98.07%
7d
98.30%
30d
98.30%
p50
414ms
p99
1.52s
Throughput
165 tps
Context
524K
$/Mtok
$0.450 / $1.20
deepinfra/fp8fp8
BaseTen—
No data24h 98.94%
7d
97.68%
30d
97.68%
p50
377ms
p99
1.99s
Throughput
202 tps
Context
1.0M
$/Mtok
$0.500 / $1.20
baseten/fp8fp8
Together—
No data24h 98.85%
7d
94.51%
30d
94.51%
p50
446ms
p99
6.20s
Throughput
89 tps
Context
524K
$/Mtok
$0.500 / $1.20
together

Measured 4m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.

History

2 endpoints reporting no data are omitted from the charts.

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.

Throughput

Median tokens per second.