# Azure

HQ US · serves 19 tracked open-weights models, 1 confirmed by its own catalog. status page · privacy policy

models
19
verified
1
of overall top 10
2
cheapest on
3 models
vs best $
2.46×
ctx (median)
253k
uptime
99.5%

# models

model rank $/M in $/M out $/M cache src ✓
deepseek-v4-flash-0731 #6 $0.44 $1.32 — litellm —
deepseek-v4-pro #7 $1.91 $3.83 $0.16 openrouter —
kimi-k2.6 #15 $0.95 $4.00 — modelsdev —
kimi-k2.7-code #19 $0.95 $4.00 $0.19 modelsdev —
deepseek-v4-flash #22 $0.21 $0.56 $0.03 openrouter —
kimi-k2.5 #23 $0.60 $3.00 — modelsdev —
kimi-k2-thinking #30 $0.60 $2.50 — litellm —
deepseek-v3.2 #32 $0.58 $1.68 — modelsdev —
deepseek-r1 #55 $1.35 $5.40 — litellm —
deepseek-v3.1 #57 $1.23 $4.94 — litellm —
gpt-oss-120b #63 $0.15 $0.60 — azure ✓
deepseek-v3 #77 $1.14 $4.56 — litellm —
mistral-large-3 #83 $0.50 $1.50 — litellm —
llama-3.3-70b-instruct #94 $0.71 $0.71 — modelsdev —
phi-4 #107 $0.12 $0.50 — modelsdev —
deepseek-v4.1-flash — $0.30 $1.20 $0.01 modelsdev —
phi-3.5-mini-instruct — $0.13 $0.52 — litellm —
phi-4-multimodal-instruct — $0.08 $0.32 — litellm —
phi-4-reasoning-plus — $0.12 $0.50 — modelsdev —

all providers · machine-readable: providers.json