gpudiff the public record of change in GPU cloud pricing
Follow the diffs: RSS · Free API · Sponsor this site

OpenRouter vs DeepInfra: LLM API price comparison

Across 67 models both gateways sell, OpenRouter is cheaper on 21, DeepInfra on 14, and 32 are priced identically. Input prices are USD per million tokens, refreshed continually; every model links to its full cross-gateway history.

ModelOpenRouter inDeepInfra in OpenRouter outDeepInfra outCheaper
claude-fable-5$10.00$10.00$50.00$50.00same
claude-opus-5$5.00$5.00$25.00$25.00same
claude-sonnet-5$2.00$3.00$10.00$15.00OpenRouter −50% →
deepseek-r1-0528$0.50$0.50$2.15$2.15same
deepseek-v3.2$0.28$0.26$0.42$0.38DeepInfra −7% →
deepseek-v4-flash$0.03$0.09$0.06$0.18OpenRouter −221% →
deepseek-v4-flash-0731$0.02$0.06$1.28$0.18OpenRouter −295% →
deepseek-v4-flash-vision-exp$0.22$0.44$0.65$1.32OpenRouter −104% →
deepseek-v4-pro$0.21$1.30$0.42$2.60OpenRouter −523% →
deepseek-v4-pro-0813$0.22$1.30$4.20$2.60OpenRouter −491% →
deepseek-v4.1-flash$0.00$0.20$2.40$0.60OpenRouter −6567% →
gemini-2.5-flash$0.30$0.30$2.50$2.50same
gemini-2.5-pro$1.25$1.25$10.00$10.00same
gemini-3.1-flash-lite$0.25$0.25$1.50$1.50same
gemini-3.5-flash$1.50$1.50$9.00$9.00same
gemini-3.7-flash$0.75$0.75$3.75$3.75same
gemma-3-12b-it$0.05$0.05$0.15$0.15same
gemma-3-27b-it$0.08$0.08$0.45$0.16same
gemma-3-4b-it$0.05$0.05$0.10$0.10same
gemma-4-26b-a4b-it$0.07$0.07$0.23$0.34OpenRouter −4% →
gemma-4-31b-it$0.09$0.15$0.34$0.40OpenRouter −67% →
glm-4.6$0.43$0.50$1.75$2.00OpenRouter −16% →
glm-4.7$0.60$0.40$2.20$1.75DeepInfra −33% →
glm-5.2$0.38$0.75$3.49$2.40OpenRouter −97% →
glm-5.3$1.40$0.90$4.40$4.00DeepInfra −36% →
glm-5.3-flash$0.15$0.15$0.50$0.50same
gpt-oss-120b$0.04$0.04$0.17$0.17same
gpt-oss-20b$0.02$0.03$0.09$0.14OpenRouter −67% →
granite-4.2-8b$0.06$0.06$0.25$0.25same
hermes-3-llama-3.1-405b$1.00$1.00$1.00$1.00same
hermes-3-llama-3.1-70b$0.70$0.70$0.70$0.70same
hy4-preview$0.83$0.83$2.50$2.50same
inkling$0.95$0.95$4.05$4.05same
inkling-small$0.45$0.45$1.20$1.20same
kimi-k2.6$0.95$0.75$4.00$3.50DeepInfra −21% →
kimi-k3$0.72$2.85$13.00$14.25OpenRouter −296% →
ling-3.0-flash$0.02$0.06$0.06$0.18OpenRouter −186% →
ling-3.0-flash-fin$0.04$0.06$0.12$0.18OpenRouter −43% →
ling-3.0-flash-vl$0.02$0.06$0.06$0.18OpenRouter −186% →
llama-guard-4-12b$0.18$0.18$0.18$0.18same
mimo-v2.6-flash$0.14$0.14$0.28$0.28same
mimo-v2.6-pro$0.43$0.43$0.87$0.87DeepInfra −1% →
minimax-m3$0.30$0.28$1.20$1.10DeepInfra −7% →
mistral-small-24b-instruct-2501$0.05$0.05$0.08$0.08same
muse-glimmer-30b$0.35$0.30$1.50$1.20DeepInfra −14% →
mythomax-l2-13b$0.08$0.40$0.11$0.40OpenRouter −400% →
phi-4$0.07$0.07$0.14$0.14same
qwen3-14b$0.12$0.12$0.24$0.24same
qwen3-30b-a3b$0.12$0.12$0.50$0.50same
qwen3-32b$0.08$0.08$0.28$0.28same
qwen3-max$0.78$1.20$3.90$6.00OpenRouter −54% →
qwen3-max-thinking$0.78$1.20$3.90$6.00OpenRouter −54% →
qwen3-next-80b-a3b-instruct$0.10$0.09$1.10$1.10DeepInfra −10% →
qwen3-vl-235b-a22b-instruct$0.21$0.20$1.90$0.88DeepInfra −5% →
qwen3-vl-30b-a3b-instruct$0.15$0.15$0.60$0.60same
qwen3.5-27b$0.20$0.26$1.56$2.60OpenRouter −33% →
qwen3.5-35b-a3b$0.15$0.14$1.00$1.00DeepInfra −7% →
qwen3.5-397b-a17b$0.55$0.45$3.50$3.00DeepInfra −18% →
qwen3.5-9b$0.10$0.10$0.15$0.15same
qwen3.6-27b$0.32$0.32$3.20$3.20same
qwen3.6-35b-a3b$0.15$0.10$1.00$0.95DeepInfra −33% →
qwen3.7-max$1.48$2.50$4.42$7.50OpenRouter −69% →
qwen3.8-2.4t-a95b$2.00$2.00$6.00$6.00same
qwen3.8-27b$0.42$0.20$3.00$2.50DeepInfra −52% →
qwen3.8-flash$0.15$0.11$0.47$0.38DeepInfra −25% →
seed-2.0-code$0.50$0.50$3.00$3.00same
seed-2.0-mini$0.10$0.10$0.40$0.40same

Fees beyond the token price

Quoted from each gateway's own documentation — we track published token prices, not invoices, so treat fee terms as their claims, not our measurement.

← All LLM prices