Qwen3.6 35B A3B: providers compared

Every endpoint serving this model, side by side. ◆ marks the best value in each column — per model, never fleet-wide.

Current state

	State
AkashMLakashml/fp8fp8	Healthy	100%◆ (best)	99.97%	99.76%	99.80%	578ms	2.93s	39 tps	262K	$0.100 / $0.900
AtlasCloudatlas-cloud/fp8fp8	Healthy	100%◆ (best)	99.27%	98.27%	98.68%	1.23s	11.2s	97 tps	262K	$0.186 / $1.11
CoreWeavecoreweave/fp8fp8	Healthy	100%◆ (best)	99.95%	99.90%	98.99%	682ms	9.31s	114 tps	262K	$0.250 / $1.25
Darkbloomdarkbloom/fp4fp4	Healthy	100%◆ (best)	99.94%	99.91%	99.84%	1.26s	11.7s	68 tps	262K	$0.050 / $0.700◆ (best)
DeepInfradeepinfra/fp8fp8	Healthy	100%◆ (best)	99.74%	99.03%	96.13%	684ms	16.5s	15 tps	262K	$0.100 / $0.950
DekaLLMdekallm	Healthy	100%◆ (best)	99.65%	99.63%	99.24%	1.02s	9.42s	67 tps	262K	$0.100 / $1.00
SiliconFlowsiliconflow/fp8fp8	Healthy	100%◆ (best)	98.95%	98.87%	98.62%	1.54s	12.6s	61 tps	262K	$0.240 / $1.80
Venicevenice/fp8fp8	Healthy	100%◆ (best)	99.62%	99.76%	99.84%	714ms	2.59s◆ (best)	136 tps◆ (best)	256K	$0.100 / $1.00
Parasailparasail/fp8fp8	Healthy	99.96%	99.37%	99.86%	99.76%	545ms◆ (best)	2.95s	43 tps	262K	$0.150 / $1.00
Darkbloomdarkbloom	No data	—	—	—	—	—	—	—	262K	$0.080 / $0.750
Io Netio-net/fp8fp8	No data	—	—	—	95.18%	—	—	—	262K	$0.220 / $1.20
Phalaphala	Down	—	97.32%	99.18%	99.43%	5.22s	23.5s	9.0 tps	262K	$0.200 / $1.27
Rekareka	No data	—	—	100%◆ (best)	99.91%◆ (best)	—	—	—	262K	$0.100 / $1.00

AkashML100%◆ (best)
Healthy24h 99.97%
7d
99.76%
30d
99.80%
p50
578ms
p99
2.93s
Throughput
39 tps
Context
262K
$/Mtok
$0.100 / $0.900
akashml/fp8fp8
AtlasCloud100%◆ (best)
Healthy24h 99.27%
7d
98.27%
30d
98.68%
p50
1.23s
p99
11.2s
Throughput
97 tps
Context
262K
$/Mtok
$0.186 / $1.11
atlas-cloud/fp8fp8
CoreWeave100%◆ (best)
Healthy24h 99.95%
7d
99.90%
30d
98.99%
p50
682ms
p99
9.31s
Throughput
114 tps
Context
262K
$/Mtok
$0.250 / $1.25
coreweave/fp8fp8
Darkbloom100%◆ (best)
Healthy24h 99.94%
7d
99.91%
30d
99.84%
p50
1.26s
p99
11.7s
Throughput
68 tps
Context
262K
$/Mtok
$0.050 / $0.700◆ (best)
darkbloom/fp4fp4
DeepInfra100%◆ (best)
Healthy24h 99.74%
7d
99.03%
30d
96.13%
p50
684ms
p99
16.5s
Throughput
15 tps
Context
262K
$/Mtok
$0.100 / $0.950
deepinfra/fp8fp8
DekaLLM100%◆ (best)
Healthy24h 99.65%
7d
99.63%
30d
99.24%
p50
1.02s
p99
9.42s
Throughput
67 tps
Context
262K
$/Mtok
$0.100 / $1.00
dekallm
SiliconFlow100%◆ (best)
Healthy24h 98.95%
7d
98.87%
30d
98.62%
p50
1.54s
p99
12.6s
Throughput
61 tps
Context
262K
$/Mtok
$0.240 / $1.80
siliconflow/fp8fp8
Venice100%◆ (best)
Healthy24h 99.62%
7d
99.76%
30d
99.84%
p50
714ms
p99
2.59s◆ (best)
Throughput
136 tps◆ (best)
Context
256K
$/Mtok
$0.100 / $1.00
venice/fp8fp8
Parasail99.96%
Healthy24h 99.37%
7d
99.86%
30d
99.76%
p50
545ms◆ (best)
p99
2.95s
Throughput
43 tps
Context
262K
$/Mtok
$0.150 / $1.00
parasail/fp8fp8
Darkbloom—
No data24h —
7d
—
30d
—
p50
—
p99
—
Throughput
—
Context
262K
$/Mtok
$0.080 / $0.750
darkbloom
Io Net—
No data24h —
7d
—
30d
95.18%
p50
—
p99
—
Throughput
—
Context
262K
$/Mtok
$0.220 / $1.20
io-net/fp8fp8
Phala—
Down24h 97.32%
7d
99.18%
30d
99.43%
p50
5.22s
p99
23.5s
Throughput
9.0 tps
Context
262K
$/Mtok
$0.200 / $1.27
phala
Reka—
No data24h —
7d
100%◆ (best)
30d
99.91%◆ (best)
p50
—
p99
—
Throughput
—
Context
262K
$/Mtok
$0.100 / $1.00
reka

Measured 2m ago. Latency and throughput are OpenRouter's rolling 30-minute windows. ◆ marks the best value in each column among these endpoints.

Price vs latency

One dot per endpoint at the latest round · further left = faster, lower = cheaper · hover a dot for its tag

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.

Throughput

Median tokens per second.