# InferenceNet

HQ US · serves 6 tracked open-weights models. status page · privacy policy

models
6
verified
—
of overall top 10
—
cheapest on
—
vs best $
2.52×
ctx (median)
1044k
uptime
100.0%

# models

model rank $/M in $/M out $/M cache src ✓
deepseek-v4.1-flash — $0.10 $0.60 $0.01 openrouter —
deepseek-v4.1-flash-1017 — $0.08 $0.55 $0.01 openrouter —
glm-5.3 — $0.08 $3.60 $0.08 openrouter —
glm-5.3-flash — $0.11 $0.35 $0.03 openrouter —
kimi-k3 — $0.23 $15.00 $0.20 openrouter —
minimax-m3.1-flash — $0.15 $0.60 $0.01 openrouter —

# recent changes

when (UTC)modelchange
Nov 1 06:26 kimi-k3 $/M output $6.97 → $15.00 ↑ 115.1%
Oct 31 21:45 deepseek-v4.1-flash $/M input $0.17 → $0.10 ↓ 40.8%
Oct 31 21:45 deepseek-v4.1-flash $/M output $0.59 → $0.60 ↑ 2.4%
Oct 31 21:45 glm-5.3 $/M input $0.10 → $0.08 ↓ 23.1%
Oct 31 21:45 glm-5.3 $/M output $2.77 → $3.60 ↑ 30.0%
Oct 31 05:58 deepseek-v4.1-flash $/M input $0.13 → $0.17 ↑ 29.9%
Oct 31 05:58 deepseek-v4.1-flash $/M output $0.76 → $0.59 ↓ 23.1%
Oct 31 05:58 kimi-k3 $/M input $0.30 → $0.23 ↓ 23.0%
Oct 31 05:58 kimi-k3 $/M output $9.07 → $6.97 ↓ 23.1%
Oct 30 22:30 deepseek-v4.1-flash $/M output $0.51 → $0.76 ↑ 49.4%
Oct 30 22:30 glm-5.3 $/M input $0.08 → $0.10 ↑ 30.0%
Oct 30 22:30 glm-5.3 $/M output $3.60 → $2.77 ↓ 23.1%
Oct 30 22:30 kimi-k3 $/M input $0.18 → $0.30 ↑ 68.9%
Oct 30 22:30 kimi-k3 $/M output $7.50 → $9.07 ↑ 20.9%

all providers · machine-readable: providers.json