Compare models

Models

Overview

Provider
Northwind
Calder Labs
Sable AI
Context window
1Mtokens
1.05Mtokens
400Ktokens
Max output
65.54Ktokens
8.19Ktokens
16.38Ktokens
Knowledge cutoff
Mar 2026
Jan 2026
Feb 2026
Input modalities
Output modalities
Open weights

Pricing

Input
$0.10/1M
$0.075/1M
$0.30/1M
Output
$0.40/1M
$0.30/1M
$1.20/1M
Cached input
$0.025/1M
$0.019/1M
$0.075/1M
Blended (3:1)
$0.175/1M
$0.131/1M
$0.525/1M

Performance

Output speed
186tok/s
214tok/s
142tok/s
Time to first token
0.3sec
0.2sec
0.3sec

Availability

Uptime (30d)
Last 30 days: 23 operational, 4 degraded, 3 outage.
99.69% average
Last 30 days: 28 operational, 2 outage.
99.86% average
Last 30 days: 27 operational, 3 degraded.
99.92% average

Benchmarks

MMLU-Pro
Score 78.4 out of 100
78.4
Score 76.9 out of 100
76.9
Score 82.6 out of 100
82.6
HumanEval
Score 84.1 out of 100
84.1
Score 81.7 out of 100
81.7
Score 87.3 out of 100
87.3
MATH
Score 71.2 out of 100
71.2
Score 68.4 out of 100
68.4
Score 77.8 out of 100
77.8

Limits

Rate limit
4,000/min
2,000/min
3,000/min
Streaming
Function calling
Fine-tuning