Leaderboard

All MPP services, ranked by quality. Tested via real payments on Tempo. Updated weekly. New services tested on discovery.

How scores work

All scores are out of 100. Only services with at least 1 successful paid test are rated.

Overall — weighted composite
Accuracy (40%) — output quality
Reliability (30%) — % requests succeeded
Latency (20%) — speed (100=<1s, 0=>30s)
Cost Eff. (10%) — quality per dollar spent via MPP
7d Vol — USDC settled on-chain (last 7 days)
Agents — unique payers on-chain
Price — cheapest endpoint (USDC)
AllWebMediaDataAISearchInfrastructureCommunicationTravelBlockchain

Rated (14)

Click a service name to see detailed ratings and test results.

1
OpenAIAI

Endpoint POST /v1/responses+5 more

100.0

High accuracy (100%) — responses consistently matched expected behavior. Very reliable — 10/10 requests got a response. Median response time: 627ms (p95: 1807ms).

Acc
100
Rel
100
Lat
100
Cost
100
7d Vol
$0.00
Agents
172
Price
from $0.0001
2
Browser UseAI

Endpoint POST /api/browser-use-mpp/run-task+3 more

97.8

High accuracy (95%) — responses consistently matched expected behavior. Very reliable — 10/10 requests got a response. Median response time: 1323ms (p95: 1703ms).

Acc
95
Rel
100
Lat
99
Cost
100
7d Vol
—
Agents
—
Price
—
3
AnthropicAI

Endpoint POST /v1/messages+1 more

97.2

High accuracy (93%) — responses consistently matched expected behavior. Very reliable — 10/10 requests got a response. Median response time: 672ms (p95: 1494ms).

Acc
93
Rel
100
Lat
100
Cost
100
7d Vol
$0.00
Agents
172
Price
from $0.020
4
OpenRouterAI

Endpoint POST /v1/chat/completions

97.2

High accuracy (93%) — responses consistently matched expected behavior. Very reliable — 10/10 requests got a response. Median response time: 655ms (p95: 1372ms).

Acc
93
Rel
100
Lat
100
Cost
100
7d Vol
$0.00
Agents
172
Price
from $0.010
5
Diffbot NLAI

Endpoint POST /diffbot-nl/analyzeOpenAPI

94.0

High accuracy (85.5%) — responses consistently matched expected behavior. Very reliable — 10/10 requests got a response. Median response time: 1225ms (p95: 1678ms).

Acc
86
Rel
100
Lat
99
Cost
100
7d Vol
—
Agents
—
Price
—
6
PerplexityAI

Endpoint POST /perplexity/chat+3 moreOpenAPI

93.3

High accuracy (85%) — responses consistently matched expected behavior. Very reliable — 10/10 requests got a response. Median response time: 1949ms (p95: 2882ms).

Acc
85
Rel
100
Lat
97
Cost
100
7d Vol
—
Agents
—
Price
—
7
GrokAI

Endpoint POST /grok/chat+6 moreOpenAPI

83.3

Moderate accuracy (72%) — some responses were incomplete or partially correct. Very reliable — 9/10 requests got a response. Median response time: 4623ms (p95: 30003ms).

Acc
72
Rel
90
Lat
88
Cost
100
7d Vol
—
Agents
—
Price
—
8
Stability AIAI

Endpoint POST /stability-ai/generate-ultra+22 moreOpenAPI

83.2

Moderate accuracy (58%) — some responses were incomplete or partially correct. Very reliable — 10/10 requests got a response. Median response time: 204ms (p95: 10982ms).

Acc
58
Rel
100
Lat
100
Cost
100
7d Vol
—
Agents
—
Price
from $23000.00
9
SunoAI

Endpoint POST /suno/generate-music+3 moreOpenAPI

80.4

Moderate accuracy (51%) — some responses were incomplete or partially correct. Very reliable — 10/10 requests got a response. Median response time: 216ms (p95: 1582ms).

Acc
51
Rel
100
Lat
100
Cost
100
7d Vol
$0.00
Agents
0
Price
from $5000.00
10
DeepSeekAI

Endpoint POST /deepseek/chat+2 moreOpenAPI

73.6

Moderate accuracy (42.22%) — some responses were incomplete or partially correct. Mostly reliable — 8/9 requests responded, 1 timed out or failed. Median response time: 65ms (p95: 3004ms).

Acc
42
Rel
89
Lat
100
Cost
100
7d Vol
$0.00
Agents
0
Price
from $3000.00
11
Mistral AIAI

Endpoint POST /mistral/chat+3 moreOpenAPI

72.0

Note: 3 test(s) excluded from accuracy — failures matched test-input error patterns, not service issues. Low accuracy (30%). Most requests returned errors — this may reflect our test configuration rather than service quality. We're working to improve coverage. Very reliable — 6/9 requests got a response. Median response time: 77ms (p95: 234ms).

Acc
30
Rel
100
Lat
100
Cost
100
7d Vol
$0.00
Agents
0
Price
from $5000.00
12
ReplicateAI

Endpoint POST /replicate/run+3 moreOpenAPI

72.0

Low accuracy (30%). Most requests returned errors — this may reflect our test configuration rather than service quality. We're working to improve coverage. Very reliable — 9/9 requests got a response. Median response time: 78ms (p95: 234ms).

Acc
30
Rel
100
Lat
100
Cost
100
7d Vol
$0.00
Agents
0
Price
from $3000.00
13
GroqAI

Endpoint POST /groq/chat+1 moreOpenAPI

72.0

Low accuracy (30%). Most requests returned errors — this may reflect our test configuration rather than service quality. We're working to improve coverage. Very reliable — 9/9 requests got a response. Median response time: 63ms (p95: 199ms).

Acc
30
Rel
100
Lat
100
Cost
100
7d Vol
$0.00
Agents
0
Price
from $5000.00
14
MathpixAI

Endpoint POST /mathpix/process-image+1 moreOpenAPI

71.7

Low accuracy (30%). Very reliable — 10/10 requests got a response. Median response time: 1442ms (p95: 1587ms).

Acc
30
Rel
100
Lat
98
Cost
100
7d Vol
—
Agents
—
Price
from $5000.00

Endpoint not available (4)

Registered in the MPP directory but not responding to requests. We check every 6 hours.

Auto.exchangeAI
Not available at last check
DiffbotAI
Not available at last check
Google GeminiAI
Not available at last check
PapercutAI
Not available at last check

14 of 18 services rated. Tests run via real MPP payments on Tempo.