MiniMax M3: 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
AtlasCloudatlas-cloud/fp8fp8	Healthy	100%◆ (best)	99.67%	99.54%	99.54%	1.15s	44.3s	78 tps	524K	$0.300 / $1.20
CoreWeavecoreweave/fp4fp4	Healthy	100%◆ (best)	99.68%	97.46%	97.46%	494ms◆ (best)	2.87s◆ (best)	124 tps	262K	$0.230 / $0.960◆ (best)
Minimaxminimax/fp8fp8	Healthy	100%◆ (best)	99.50%	99.64%	99.64%	1.26s	13.5s	68 tps	524K	$0.300 / $1.20
ModelRunmodelrun/fp4fp4	Healthy	100%◆ (best)	99.80%	99.77%	99.77%	1.80s	13.4s	93 tps	1.0M	$0.750 / $3.00
Morphmorph/fp4fp4	Healthy	100%◆ (best)	96.53%	97.44%	97.44%	2.06s	16.3s	25 tps	256K	$0.255 / $1.02
Venicevenice/fp8fp8	Healthy	100%◆ (best)	99.88%	99.82%◆ (best)	99.82%◆ (best)	1.10s	7.91s	116 tps	524K	$0.300 / $1.20
Novitanovita/fp8fp8	Healthy	99.90%	99.60%	99.66%	99.66%	2.24s	23.9s	52 tps	1.0M	$0.300 / $1.20
Togethertogether	Healthy	99.76%	99.66%	99.77%	99.77%	1.12s	10.7s	59 tps	524K	$0.300 / $1.20
GMICloudgmicloud/fp8fp8	Healthy	99.18%	99.49%	99.36%	99.36%	3.26s	10.2s	38 tps	1.0M	$0.240 / $0.960
StreamLakestreamlake/fp8fp8	Healthy	98.98%	98.96%	98.47%	98.47%	1.61s	7.73s	56 tps	1.0M	$0.300 / $1.20
Parasailparasail/fp8fp8	Healthy	97.78%	98.19%	97.89%	97.89%	1.12s	22.7s	47 tps	1.0M	$0.300 / $1.20
DeepInfradeepinfra/fp8fp8	Down	81.20%	98.22%	98.36%	98.36%	783ms	9.49s	27 tps	524K	$0.280 / $1.10
SambaNovasambanova	No data	—	98.03%	98.60%	98.60%	1.98s	8.02s	135 tps◆ (best)	1.0M	$0.600 / $2.40

AtlasCloud100%◆ (best)
Healthy24h 99.67%
7d
99.54%
30d
99.54%
p50
1.15s
p99
44.3s
Throughput
78 tps
Context
524K
$/Mtok
$0.300 / $1.20
atlas-cloud/fp8fp8
CoreWeave100%◆ (best)
Healthy24h 99.68%
7d
97.46%
30d
97.46%
p50
494ms◆ (best)
p99
2.87s◆ (best)
Throughput
124 tps
Context
262K
$/Mtok
$0.230 / $0.960◆ (best)
coreweave/fp4fp4
Minimax100%◆ (best)
Healthy24h 99.50%
7d
99.64%
30d
99.64%
p50
1.26s
p99
13.5s
Throughput
68 tps
Context
524K
$/Mtok
$0.300 / $1.20
minimax/fp8fp8
ModelRun100%◆ (best)
Healthy24h 99.80%
7d
99.77%
30d
99.77%
p50
1.80s
p99
13.4s
Throughput
93 tps
Context
1.0M
$/Mtok
$0.750 / $3.00
modelrun/fp4fp4
Morph100%◆ (best)
Healthy24h 96.53%
7d
97.44%
30d
97.44%
p50
2.06s
p99
16.3s
Throughput
25 tps
Context
256K
$/Mtok
$0.255 / $1.02
morph/fp4fp4
Venice100%◆ (best)
Healthy24h 99.88%
7d
99.82%◆ (best)
30d
99.82%◆ (best)
p50
1.10s
p99
7.91s
Throughput
116 tps
Context
524K
$/Mtok
$0.300 / $1.20
venice/fp8fp8
Novita99.90%
Healthy24h 99.60%
7d
99.66%
30d
99.66%
p50
2.24s
p99
23.9s
Throughput
52 tps
Context
1.0M
$/Mtok
$0.300 / $1.20
novita/fp8fp8
Together99.76%
Healthy24h 99.66%
7d
99.77%
30d
99.77%
p50
1.12s
p99
10.7s
Throughput
59 tps
Context
524K
$/Mtok
$0.300 / $1.20
together
GMICloud99.18%
Healthy24h 99.49%
7d
99.36%
30d
99.36%
p50
3.26s
p99
10.2s
Throughput
38 tps
Context
1.0M
$/Mtok
$0.240 / $0.960
gmicloud/fp8fp8
StreamLake98.98%
Healthy24h 98.96%
7d
98.47%
30d
98.47%
p50
1.61s
p99
7.73s
Throughput
56 tps
Context
1.0M
$/Mtok
$0.300 / $1.20
streamlake/fp8fp8
Parasail97.78%
Healthy24h 98.19%
7d
97.89%
30d
97.89%
p50
1.12s
p99
22.7s
Throughput
47 tps
Context
1.0M
$/Mtok
$0.300 / $1.20
parasail/fp8fp8
DeepInfra81.20%
Down24h 98.22%
7d
98.36%
30d
98.36%
p50
783ms
p99
9.49s
Throughput
27 tps
Context
524K
$/Mtok
$0.280 / $1.10
deepinfra/fp8fp8
SambaNova—
No data24h 98.03%
7d
98.60%
30d
98.60%
p50
1.98s
p99
8.02s
Throughput
135 tps◆ (best)
Context
1.0M
$/Mtok
$0.600 / $2.40
sambanova

Measured 4m 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.