The Known Good Updated 23 Oct 2026 Subscribe
Providers· Inference host· 12 tracked models

Makora — pricing & performance

Compare models All providers Data sources
Models served
12
of 1,298 tracked models
Mean output speed
94
tokens per second, averaged over
11 measured endpoints
Mean latency
0.8s
time to first token, averaged over
11 measured endpoints
Cheapest blended
$0.11
per 1M tokens at a 3:1
input:output blend
Output speed on Makora

Output tokens per second for the 11 of 12 models on Makora 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.

⇩
152
128
112
105
101
96
80
78
75
59
44
The Known Good
12 models on Makora
Full leaderboard →
Model Creator Input $/M Output $/M Blended Speed TTFT Context
DeepSeek: DeepSeek V4 Flash 0731 DeepSeek $0.090 $0.20 $0.12 44 0.6s 1.00M
DeepSeek: DeepSeek V4.1 Flash DeepSeek $0.26 $1.17 $0.49 152 0.5s 1.05M
DeepSeek: DeepSeek V4.1 Flash 1017 DeepSeek $0.15 $1.20 $0.41 105 1.1s 1.05M
Google: Gemma 4 26B A4B Google $0.070 $0.24 $0.11 75 0.4s 256k
Meta: Muse Spark Open Meta $1.25 $4.25 $2.00 80 1.8s 1.05M
MoonshotAI: Kimi K3 Moonshot AI $0.65 $15.00 $4.24 59 0.9s 1.05M
Qwen: Qwen3.8 Flash Alibaba $0.15 $0.47 $0.23 128 0.5s 262k
Tencent: Hy4 preview Tencent $0.80 $2.48 $1.22 — — 131k
Xiaomi: MiMo-V2.6-Flash Xiaomi $0.13 $0.28 $0.17 101 0.6s 1.05M
Z.ai: GLM 5.2 Z.ai $0.72 $2.38 $1.14 78 0.9s 980k
Z.ai: GLM 5.3 Z.ai $0.14 $4.40 $1.21 112 0.8s 1.05M
Z.ai: GLM 5.3 Flash Z.ai $0.14 $0.47 $0.22 96 0.8s 1.05M
The Known Good
Source. Prices, endpoint availability and throughput for Makora 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.