The Known Good Updated 29 Jul 2026
Providers· Inference host· 14 tracked models

GMICloud — pricing & performance

Compare models All providers Data sources
Models served
14
of 947 tracked models
Mean output speed
38
tokens per second, averaged over
13 measured endpoints
Mean latency
2.6s
time to first token, averaged over
13 measured endpoints
Cheapest blended
$0.100
per 1M tokens at a 3:1
input:output blend
Output speed on GMICloud

Median output tokens per second for the 13 of 14 models on GMICloud with published throughput telemetry from OpenRouter. The full table below lists all 14, measured or not. Higher is better. We do not run these measurements ourselves.

⇩
66
48
48
44
37
37
36
35
34
33
26
22
The Known Good
14 models on GMICloud
Full leaderboard →
Model Creator Input $/M Output $/M Blended Speed TTFT Context
DeepSeek: DeepSeek V3.2 DeepSeek $0.29 $0.43 $0.33 26 2.5s 164k
DeepSeek: DeepSeek V4 Flash DeepSeek $0.094 $0.19 $0.12 37 2.2s 1.05M
DeepSeek: DeepSeek V4 Pro DeepSeek $0.68 $1.36 $0.85 48 2.5s 1.05M
MiniMax: MiniMax M2.7 MiniMax $0.30 $1.20 $0.53 34 2.0s 197k
MiniMax: MiniMax M3 MiniMax $0.24 $0.96 $0.42 33 2.0s 1.05M
MoonshotAI: Kimi K3 Moonshot AI $3.00 $15.00 $6.00 22 3.9s 262k
Qwen: Qwen3.5 397B A17B Alibaba $0.60 $3.60 $1.35 — — 262k
Tencent: Hy3 Tencent $0.13 $0.53 $0.23 48 0.9s 262k
Tencent: Hy3 preview Tencent $0.063 $0.21 $0.100 44 2.3s 262k
Xiaomi: MiMo-V2.5 Xiaomi $0.11 $0.22 $0.14 36 3.7s 1.05M
Xiaomi: MiMo-V2.5-Pro Xiaomi $0.35 $0.70 $0.43 22 4.3s 1.05M
Z.ai: GLM 5 Z.ai $0.60 $1.92 $0.93 35 1.7s 203k
Z.ai: GLM 5.1 Z.ai $0.98 $3.08 $1.50 66 2.7s 203k
Z.ai: GLM 5.2 Z.ai $0.92 $2.90 $1.42 37 3.3s 1.05M
The Known Good
Source. Prices, endpoint availability and throughput for GMICloud 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.