Skip to content

Same-model provider benchmark

Crusoe vs StreamLake: LLM provider comparison

Compare Crusoe and StreamLake on 3 exact shared text models. ProviderBench keeps speed, price, and catalog coverage separate so naturally faster model catalogs cannot distort the result.

Crusoe

7 indexed models

Headquarters
United States
Server regions
N/a
Model types
7 text

StreamLake

20 indexed models

Headquarters
China
Server regions
N/a
Model types
20 text

At a glance

Metric winners

There is no overall score. Each winner answers one specific question using only directly comparable data.

Fastest on shared models

1.02× typical advantage

3 exact models with complete recent speed data

Lowest token cost

$1.282 / 1M

3 exact models using a 1K-input/500-output mix

Most models available

20 models

Complete catalog coverage across all indexed modalities

Shared-model benchmark summary

MetricCrusoe(0)StreamLake(0)
SpeedCrusoe11.96 sStreamLake17.23 s
TTFTCrusoe0.39 sStreamLake1.51 s
TPSCrusoe43.0 tok/sStreamLake32.0 tok/s
UptimeCrusoe99.58%StreamLake99.48%
BlendedCrusoe$1.4378 / 1MStreamLake$1.282 / 1M
Green value Better comparable resultRed value Worse comparable result

Visual comparison

Price and performance charts

CrusoeStreamLake
500-token response by shared model

Estimated seconds using recent median response-start and output-speed data. Lower is better.

StreamLake has the lower typical same-model response ratio across 3 measured models.

Blended token price by shared model

USD per 1 million tokens using a 1,000-input/500-output mix. Lower is better.

StreamLake has the lower typical price ratio across 3 priced shared models.

Shared text models

3 exact models · newest first

ModelCrusoe(0)StreamLake(0)
Kimi K2.6Winner · StreamLake
Crusoe
Speed— Worst comparable value
34.64 s
TTFT— Best comparable value
1.31 s
TPS— Worst comparable value
15.0 tok/s
Uptime— Worst comparable value
56.00%
Context
262.1K
Route
crusoe/bf16
Blended— Best comparable value
$1.6333 / 1M
StreamLake
Speed— Best comparable value
18.17 s
TTFT— Worst comparable value
1.51 s
TPS— Best comparable value
30.0 tok/s
Uptime— Best comparable value
99.48%
Context
256K
Route
streamlake/fp8
Blended— Worst comparable value
$1.77 / 1M
GLM 5.1Winner · Crusoe
Crusoe
Speed— Best comparable value
5.89 s
TTFT— Best comparable value
0.39 s
TPS— Best comparable value
91.0 tok/s
Uptime— Best comparable value
100.00%
Context
202.8K
Route
crusoe/fp8
Blended— Worst comparable value
$2.2667 / 1M
StreamLake
Speed— Worst comparable value
17.23 s
TTFT— Worst comparable value
1.60 s
TPS— Worst comparable value
32.0 tok/s
Uptime— Worst comparable value
99.43%
Context
200K
Route
streamlake/fp8
Blended— Best comparable value
$1.656 / 1M
Qwen3 235B A22B Instruct 2507Winner · StreamLake
Crusoe
Speed— Worst comparable value
11.96 s
TTFT— Best comparable value
0.33 s
TPS— Worst comparable value
43.0 tok/s
Uptime— Worst comparable value
99.58%
Context
262.1K
Route
crusoe/bf16
Blended— Best comparable value
$0.4133 / 1M
StreamLake
Speed— Best comparable value
11.77 s
TTFT— Worst comparable value
0.66 s
TPS— Best comparable value
45.0 tok/s
Uptime— Best comparable value
99.71%
Context
128K
Route
streamlake
Blended— Worst comparable value
$0.42 / 1M

A per-model winner combines blended price and estimated 500-token response time with equal proportional weight. Ties and rows missing either measurement receive no badge. Comparison data calculated . Values use one deterministic route per provider and model; missing measurements remain visible as N/a.

Crusoe vs StreamLake analysis

How Crusoe and StreamLake compare for AI inference

Crusoe and StreamLake share 3 indexed text models, including Kimi K2.6, GLM 5.1, Qwen3 235B A22B Instruct 2507. StreamLake has the stronger typical response-time result on the directly measured set.

Same-model speed evidence

3 shared models currently have complete response-start and output-speed measurements on both providers. The speed comparison uses per-model ratios before taking the median, so naturally faster model catalogs do not improve the result.

Token pricing on one workload

3 shared models have complete input and output prices on both providers. Prices use the same 1,000-input/500-output-token mix and are normalized to one million tokens for readability.

Catalog and deployment differences

Crusoe has 7 indexed models and lists no published server regions; StreamLake has 20 models and lists no regions. Verify data residency, privacy terms, limits, and production latency directly before choosing.