Skip to content

Same-model provider benchmark

CoreWeave vs Wafer: LLM provider comparison

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

CoreWeave

20 indexed models

Headquarters
United States
Server regions
United States
Model types
20 text

Wafer

3 indexed models

Headquarters
United States
Server regions
N/a
Model types
3 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

1.69× typical advantage

3 exact models with complete recent speed data

Lowest token cost

$1.9111 / 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

MetricCoreWeave(0)Wafer(0)
SpeedCoreWeave19.86 sWafer14.75 s
TTFTCoreWeave1.13 sWafer4.95 s
TPSCoreWeave27.0 tok/sWafer53.0 tok/s
UptimeCoreWeave99.37%Wafer100.00%
BlendedCoreWeave$2.0111 / 1MWafer$1.9111 / 1M
Green value Better comparable resultRed value Worse comparable result

Visual comparison

Price and performance charts

CoreWeaveWafer
500-token response by shared model

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

Wafer 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.

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

Shared text models

3 exact models · newest first

ModelCoreWeave(0)Wafer(0)
GLM 5.2Winner · Wafer
CoreWeave
Speed— Worst comparable value
19.86 s
TTFT— Best comparable value
1.35 s
TPS— Worst comparable value
27.0 tok/s
Uptime— Worst comparable value
99.62%
Context
262.1K
Route
coreweave/fp4
Blended— Best comparable value
$1.3133 / 1M
Wafer
Speed— Best comparable value
9.16 s
TTFT— Worst comparable value
2.41 s
TPS— Best comparable value
74.0 tok/s
Uptime— Best comparable value
100.00%
Context
1M
Route
wafer/fp4
Blended— Worst comparable value
$2.4 / 1M
DeepSeek V4 ProWinner · Wafer
CoreWeave
Speed— Worst comparable value
24.94 s
TTFT— Best comparable value
1.13 s
TPS— Worst comparable value
21.0 tok/s
Uptime— Worst comparable value
97.76%
Context
1M
Route
coreweave/fp8
Blended— Worst comparable value
$2.32 / 1M
Wafer
Speed— Best comparable value
14.75 s
TTFT— Worst comparable value
4.95 s
TPS— Best comparable value
51.0 tok/s
Uptime— Best comparable value
99.28%
Context
1M
Route
wafer/fp4
Blended— Best comparable value
$1.6 / 1M
GLM 5.1Winner · CoreWeave
CoreWeave
Speed— Best comparable value
4.42 s
TTFT— Best comparable value
0.51 s
TPS— Best comparable value
128.0 tok/s
Uptime— Worst comparable value
99.37%
Context
202.8K
Route
coreweave/fp8
Blended— Worst comparable value
$2.4 / 1M
Wafer
Speed— Worst comparable value
15.15 s
TTFT— Worst comparable value
5.71 s
TPS— Worst comparable value
53.0 tok/s
Uptime— Best comparable value
100.00%
Context
202.8K
Route
wafer/fp4
Blended— Best comparable value
$1.7333 / 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.

CoreWeave vs Wafer analysis

How CoreWeave and Wafer compare for AI inference

CoreWeave and Wafer share 3 indexed text models, including GLM 5.2, DeepSeek V4 Pro, GLM 5.1. Wafer 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

CoreWeave has 20 indexed models and lists 1 published server region; Wafer has 3 models and lists no regions. Verify data residency, privacy terms, limits, and production latency directly before choosing.