Skip to content

Same-model provider benchmark

Alibaba Cloud Int. vs CoreWeave: LLM provider comparison

Compare Alibaba Cloud Int. and CoreWeave on 7 exact shared text models. ProviderBench keeps speed, price, and catalog coverage separate so naturally faster model catalogs cannot distort the result.

Alibaba Cloud Int.

50 indexed models

Headquarters
Singapore
Server regions
SingaporeChina
Model types
45 text, 2 video, 2 speech, 1 transcription

CoreWeave

20 indexed models

Headquarters
United States
Server regions
United States
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

1.31× typical advantage

7 exact models with complete recent speed data

Lowest token cost

$1.1085 / 1M

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

Most models available

50 models

Complete catalog coverage across all indexed modalities

Shared-model benchmark summary

MetricAlibaba Cloud Int.(0)CoreWeave(0)
SpeedAlibaba Cloud Int.10.50 sCoreWeave17.27 s
TTFTAlibaba Cloud Int.0.97 sCoreWeave0.35 s
TPSAlibaba Cloud Int.58.0 tok/sCoreWeave30.0 tok/s
UptimeAlibaba Cloud Int.99.92%CoreWeave99.98%
BlendedAlibaba Cloud Int.$1.1085 / 1MCoreWeave$1.2024 / 1M
Green value Better comparable resultRed value Worse comparable result

Visual comparison

Price and performance charts

Alibaba Cloud Int.CoreWeave
500-token response by shared model

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

Alibaba Cloud Int. has the lower typical same-model response ratio across 7 measured models.

Blended token price by shared model

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

Alibaba Cloud Int. has the lower typical price ratio across 7 priced shared models.

Shared text models

7 exact models · newest first

ModelAlibaba Cloud Int.(0)CoreWeave(0)
GLM 5.2Winner · Alibaba Cloud Int.
Alibaba Cloud Int.
Speed— Best comparable value
12.95 s
TTFT— Best comparable value
1.33 s
TPS— Best comparable value
43.0 tok/s
Uptime— Worst comparable value
99.91%
Context
1M
Route
alibaba
Blended— Best comparable value
$1.416 / 1M
CoreWeave
Speed— Worst comparable value
20.46 s
TTFT— Worst comparable value
1.94 s
TPS— Worst comparable value
27.0 tok/s
Uptime— Best comparable value
99.96%
Context
262.1K
Route
coreweave/fp4
Blended— Worst comparable value
$2.3933 / 1M
Qwen3.6 27BWinner · Alibaba Cloud Int.
Alibaba Cloud Int.
Speed— Best comparable value
14.42 s
TTFT— Worst comparable value
0.91 s
TPS— Best comparable value
37.0 tok/s
Uptime— Worst comparable value
99.91%
Context
262.1K
Route
alibaba
Blended— Best comparable value
$1.2 / 1M
CoreWeave
Speed— Worst comparable value
18.18 s
TTFT— Best comparable value
0.33 s
TPS— Worst comparable value
28.0 tok/s
Uptime— Best comparable value
100.00%
Context
262.1K
Route
coreweave/fp8
Blended— Worst comparable value
$1.6 / 1M
DeepSeek V4 ProWinner · Alibaba Cloud Int.
Alibaba Cloud Int.
Speed— Best comparable value
10.50 s
TTFT— Worst comparable value
1.88 s
TPS— Best comparable value
58.0 tok/s
Uptime— Best comparable value
99.11%
Context
1M
Route
alibaba
Blended— Best comparable value
$1.888 / 1M
CoreWeave
Speed— Worst comparable value
64.00 s
TTFT— Best comparable value
1.50 s
TPS— Worst comparable value
8.0 tok/s
Uptime— Worst comparable value
98.59%
Context
1M
Route
coreweave/fp8
Blended— Worst comparable value
$2.32 / 1M
DeepSeek V4 FlashWinner · Alibaba Cloud Int.
Alibaba Cloud Int.
Speed— Best comparable value
8.02 s
TTFT— Worst comparable value
0.97 s
TPS— Best comparable value
71.0 tok/s
Uptime— Best comparable value
99.93%
Context
1M
Route
alibaba
Blended— Best comparable value
$0.1787 / 1M
CoreWeave
Speed— Worst comparable value
17.27 s
TTFT— Best comparable value
0.60 s
TPS— Worst comparable value
30.0 tok/s
Uptime— Worst comparable value
99.85%
Context
1M
Route
coreweave/fp8
Blended— Worst comparable value
$0.1867 / 1M
Qwen3.5-35B-A3BWinner · Alibaba Cloud Int.
Alibaba Cloud Int.
Speed— Best comparable value
4.71 s
TTFT— Worst comparable value
0.88 s
TPS— Best comparable value
130.5 tok/s
Uptime
100.00%
Context
262.1K
Route
alibaba
Blended— Best comparable value
$0.5417 / 1M
CoreWeave
Speed— Worst comparable value
4.86 s
TTFT— Best comparable value
0.35 s
TPS— Worst comparable value
111.0 tok/s
Uptime
100.00%
Context
262.1K
Route
coreweave/fp8
Blended— Worst comparable value
$0.5833 / 1M
Qwen3 30B A3B Instruct 2507Winner · CoreWeave
Alibaba Cloud Int.
Speed— Best comparable value
6.28 s
TTFT— Worst comparable value
0.32 s
TPS— Best comparable value
84.0 tok/s
Uptime
100.00%
Context
131.1K
Route
alibaba
Blended— Worst comparable value
$0.26 / 1M
CoreWeave
Speed— Worst comparable value
8.22 s
TTFT— Best comparable value
0.28 s
TPS— Worst comparable value
63.0 tok/s
Uptime
100.00%
Context
262.1K
Route
coreweave/bf16
Blended— Best comparable value
$0.1667 / 1M
Qwen3 Coder 480B A35BWinner · CoreWeave
Alibaba Cloud Int.
Speed— Worst comparable value
12.10 s
TTFT— Worst comparable value
1.90 s
TPS— Worst comparable value
49.0 tok/s
Uptime
N/a
Context
262.1K
Route
alibaba/opensource
Blended— Worst comparable value
$2.275 / 1M
CoreWeave
Speed— Best comparable value
9.06 s
TTFT— Best comparable value
0.29 s
TPS— Best comparable value
57.0 tok/s
Uptime
N/a
Context
262.1K
Route
coreweave/bf16
Blended— Best comparable value
$1.1667 / 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.

Alibaba Cloud Int. vs CoreWeave analysis

How Alibaba Cloud Int. and CoreWeave compare for AI inference

Alibaba Cloud Int. and CoreWeave share 7 indexed text models, including GLM 5.2, Qwen3.6 27B, DeepSeek V4 Pro, DeepSeek V4 Flash, Qwen3.5-35B-A3B. Alibaba Cloud Int. has the stronger typical response-time result on the directly measured set.

Same-model speed evidence

7 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

7 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

Alibaba Cloud Int. has 50 indexed models and lists 2 published server regions; CoreWeave has 20 models and lists 1 region. Verify data residency, privacy terms, limits, and production latency directly before choosing.