The Known Good Updated 24 Oct 2026 Subscribe
Providers· Inference host· 12 tracked models

Sail Research — pricing & performance

Compare models All providers Data sources
Models served
12
of 1,302 tracked models
Mean output speed
64
tokens per second, averaged over
10 measured endpoints
Mean latency
1.0s
time to first token, averaged over
10 measured endpoints
Cheapest blended
$0.075
per 1M tokens at a 3:1
input:output blend
Output speed on Sail Research

Output tokens per second for the 10 of 12 models on Sail Research with published throughput telemetry, the most recent figure OpenRouter publishes for each host. The full table below lists all 12, measured or not. Higher is better. We do not run these measurements ourselves.

All 10 of 10 models⇩
141
100
83
74
68
67
37
27
22
17
The Known Good
12 models on Sail Research
Full leaderboard →
Model Creator Input $/M Output $/M Blended Speed TTFT Context
DeepSeek: DeepSeek V4 Flash 0423 DeepSeek $0.090 $0.18 $0.11 22 0.8s 1.05M
DeepSeek: DeepSeek V4 Flash 0731 DeepSeek $0.15 $0.15 $0.15 74 1.0s 1.05M
DeepSeek: DeepSeek V4 Pro 0813 DeepSeek $0.45 $0.80 $0.54 100 0.8s 1.05M
DeepSeek: DeepSeek V4.1 Flash DeepSeek $0.090 $0.25 $0.13 141 0.9s 1.05M
DeepSeek: DeepSeek V4.1 Flash (batch) DeepSeek $0.050 $0.15 $0.075 — — 1.05M
DeepSeek: DeepSeek V4.1 Flash 1017 DeepSeek $0.10 $0.15 $0.11 67 1.2s 1.05M
MoonshotAI: Kimi K2.6 Moonshot AI $1.00 $4.00 $1.75 27 0.6s 262k
MoonshotAI: Kimi K3 Moonshot AI $1.20 $5.50 $2.27 83 0.7s 1.05M
Z.ai: GLM 5.2 Z.ai $0.50 $3.15 $1.16 37 1.3s 1.05M
Z.ai: GLM 5.3 Z.ai $0.50 $2.00 $0.88 68 1.1s 1.05M
Z.ai: GLM 5.3 Flash Z.ai $0.078 $0.25 $0.12 17 1.8s 1.05M
Z.ai: GLM 5.3 Flash (batch) Z.ai $0.050 $0.18 $0.083 — — 1.05M
The Known Good
Source. Prices, endpoint availability and throughput for Sail Research are ingested from OpenRouter; model metadata is enriched from models.dev. A blank cell means the figure was not published for that endpoint — it never means zero. Blended price is (3 × input + output) ÷ 4, the same 3:1 blend used everywhere on this site; see methodology. Quality scores are not provider-specific: a model scores the same wherever it is hosted, so quality lives on the model pages.