数据状态

同步自 OpenRouter Providers · 117 家托管商 · source=openrouter · 2026-10-31T00:35:16.591Z

deepinfra

DeepInfra

总部 美国

69 模型112 端点日可用 96.8%
托管模型69
端点112
日可用率均96.8%
最低输入 $/M$0.020
最大上下文1.0M

托管端点 · 112

来自 OpenRouter 各模型 Endpoints API:同一模型可能有多量化 / 多区域端点。

模型Tag量化上下文输入输出5m30m1d延迟吞吐缓存
llama-3.1-8b-instruct
meta-llama/llama-3.1-8b-instruct
deepinfra/fp8fp8131k$0.020$0.040100.00%100.00%100.00%———
mistral-nemo
mistralai/mistral-nemo
deepinfra/fp8fp8131k$0.029$0.030100.00%100.00%100.00%———
gemma-3-12b-it
google/gemma-3-12b-it
deepinfra/bf16/flexbf16131k$0.040$0.120——100.00%———
l3-lunaris-8b
sao10k/l3-lunaris-8b
deepinfra/turbofp88k$0.040$0.050100.00%100.00%100.00%———
gemma-3-12b-it
google/gemma-3-12b-it
deepinfra/bf16bf16131k$0.050$0.150100.00%100.00%100.00%———
ling-3.0-flash-vl
inclusionai/ling-3.0-flash-vl
deepinfra/fp16fp16131k$0.060$0.180100.00%100.00%100.00%——是
granite-4.2-8b
ibm-granite/granite-4.2-8b
deepinfra/bf16bf16131k$0.060$0.250100.00%100.00%100.00%——是
deepseek-v4-flash-0731
deepseek/deepseek-v4-flash-0731
deepinfra/fp8fp81.0M$0.060$0.180100.00%100.00%100.00%——是
ling-3.1-flash
inclusionai/ling-3.1-flash
deepinfra/fp4fp4262k$0.075$0.225100.00%100.00%100.00%——是
gemma-3-12b-it
google/gemma-3-12b-it
deepinfra/bf16/prioritybf16131k$0.075$0.225100.00%100.00%100.00%———
nemotron-3-super-120b-a12b
nvidia/nemotron-3-super-120b-a12b
deepinfra/bf16bf16262k$0.085$0.400100.00%100.00%100.00%———
deepseek-v4-flash
deepseek/deepseek-v4-flash
deepinfra/fp8fp81.0M$0.090$0.180100.00%100.00%100.00%——是
deepseek-v4.1-flash-1017
deepseek/deepseek-v4.1-flash-1017
deepinfra/fp8fp81.0M$0.098$0.294100.00%100.00%100.00%——是
qwen3-30b-a3b
qwen/qwen3-30b-a3b
deepinfra/fp8fp841k$0.120$0.500100.00%100.00%100.00%———
qwen3-14b
qwen/qwen3-14b
deepinfra/fp8fp841k$0.120$0.240100.00%100.00%100.00%———
qwen3-next-80b-a3b-instruct
qwen/qwen3-next-80b-a3b-instruct
deepinfra/fp8/priorityfp8262k$0.135$1.65100.00%100.00%100.00%———
deepseek-v4.1-flash
deepseek/deepseek-v4.1-flash
deepinfra/fp8fp81.0M$0.140$0.420100.00%100.00%100.00%——是
deepseek-v4.1-flash
deepseek/deepseek-v4.1-flash
deepinfra/fp8fp81.0M$0.140$0.420100.00%100.00%100.00%——是
deepseek-v4.1-flash-1017
deepseek/deepseek-v4.1-flash-1017
deepinfra/fp8/priorityfp81.0M$0.147$0.441100.00%100.00%100.00%——是
nemotron-3.5-super-vl
nvidia/nemotron-3.5-super-vl
deepinfra/fp4fp4262k$0.150$0.500100.00%100.00%100.00%——是
nemotron-3.5-super-vl
nvidia/nemotron-3.5-super-vl
deepinfra/fp4fp4262k$0.150$0.500100.00%100.00%100.00%——是
glm-5.3-flash
z-ai/glm-5.3-flash
deepinfra/fp4fp41.0M$0.150$0.500100.00%100.00%100.00%——是
glm-5.3-flash
z-ai/glm-5.3-flash
deepinfra/fp4fp41.0M$0.150$0.500100.00%100.00%100.00%——是
gemma-4-31b-it
google/gemma-4-31b-it
deepinfra/fp8/flexfp8262k$0.160$0.320——100.00%———
gemma-4-31b-it
google/gemma-4-31b-it
deepinfra/fp8/flexfp8262k$0.160$0.320——100.00%———
nemotron-3.5-content-safety
nvidia/nemotron-3.5-content-safety
deepinfra/bf16bf16131k$0.200$0.200——100.00%———
nemotron-3.5-content-safety
nvidia/nemotron-3.5-content-safety
deepinfra/bf16bf16131k$0.200$0.200——100.00%———
minimax-m3
minimax/minimax-m3
deepinfra/fp8/flexfp8524k$0.224$0.880——100.00%——是
deepseek-chat-v3.1
deepseek/deepseek-chat-v3.1
deepinfra/fp4fp4164k$0.250$0.950100.00%100.00%100.00%——是
deepseek-chat
deepseek/deepseek-chat
deepinfra/fp4/flexfp4164k$0.256$0.712——100.00%———
qwen3.5-27b
qwen/qwen3.5-27b
deepinfra/fp8fp8262k$0.260$2.60—100.00%100.00%———
minimax-m3
minimax/minimax-m3
deepinfra/fp8fp8524k$0.280$1.10100.00%100.00%100.00%——是
muse-glimmer-30b
meta/muse-glimmer-30b
deepinfra/bf16bf16131k$0.300$1.20100.00%100.00%100.00%——是
qwen3.6-27b
qwen/qwen3.6-27b
deepinfra/fp8fp8262k$0.320$3.20100.00%100.00%100.00%———
glm-4.7
z-ai/glm-4.7
deepinfra/fp4/flexfp4203k$0.320$1.40——100.00%——是
mimo-v2.6-pro
xiaomi/mimo-v2.6-pro
deepinfra/fp8fp81.0M$0.387$0.783100.00%100.00%100.00%——是
nemotron-3-ultra-550b-a55b
nvidia/nemotron-3-ultra-550b-a55b
deepinfra/fp4/flexfp4262k$0.400$1.76——100.00%——是
nemotron-3-ultra-550b-a55b
nvidia/nemotron-3-ultra-550b-a55b
deepinfra/fp4/flexfp4262k$0.400$1.76——100.00%——是
glm-4.6
z-ai/glm-4.6
deepinfra/fp4/flexfp4203k$0.400$1.60——100.00%——是
inkling-small
thinkingmachines/inkling-small
deepinfra/fp8fp8524k$0.450$1.20—100.00%100.00%——是
inkling-small
thinkingmachines/inkling-small
deepinfra/fp8fp8524k$0.450$1.20—100.00%100.00%——是
glm-5.2
z-ai/glm-5.2
deepinfra/fp4/flexfp41.0M$0.450$1.44——100.00%——是
nemotron-3-ultra-550b-a55b
nvidia/nemotron-3-ultra-550b-a55b
deepinfra/fp4fp4262k$0.500$2.20100.00%100.00%100.00%——是
nemotron-3-ultra-550b-a55b
nvidia/nemotron-3-ultra-550b-a55b
deepinfra/fp4fp4262k$0.500$2.20100.00%100.00%100.00%——是
deepseek-r1-0528
deepseek/deepseek-r1-0528
deepinfra/fp4fp4164k$0.500$2.15100.00%100.00%100.00%——是
hermes-3-llama-3.1-70b
nousresearch/hermes-3-llama-3.1-70b
deepinfra/fp8fp8131k$0.700$0.700100.00%100.00%100.00%———
hy4-preview
tencent/hy4-preview
deepinfra/fp4fp41.0M$0.792$2.38100.00%100.00%100.00%——是
l3.1-euryale-70b
sao10k/l3.1-euryale-70b
deepinfra/fp8fp8131k$0.850$0.850—100.00%100.00%———
deepseek-v4-pro-0813
deepseek/deepseek-v4-pro-0813
deepinfra/fp8fp81.0M$1.30$2.60100.00%100.00%100.00%——是
gemma-4-26b-a4b-it
google/gemma-4-26b-a4b-it
deepinfra/fp8fp8262k$0.070$0.340100.00%100.00%99.93%———
gemma-4-26b-a4b-it
google/gemma-4-26b-a4b-it
deepinfra/fp8fp8262k$0.070$0.340100.00%100.00%99.93%———
llama-3.3-70b-instruct
meta-llama/llama-3.3-70b-instruct
deepinfra/turbofp8131k$0.100$0.320100.00%100.00%99.93%———
gemma-4-31b-it
google/gemma-4-31b-it
deepinfra/turbofp4262k$0.135$0.510100.00%100.00%99.93%——是
gemma-4-31b-it
google/gemma-4-31b-it
deepinfra/turbofp4262k$0.135$0.510100.00%100.00%99.93%——是
mimo-v2.6-flash
xiaomi/mimo-v2.6-flash
deepinfra/fp8fp81.0M$0.140$0.280100.00%100.00%99.93%——是
mistral-small-24b-instruct-2501
mistralai/mistral-small-24b-instruct-2501
deepinfra/fp8fp833k$0.050$0.080100.00%100.00%99.93%———
phi-4
microsoft/phi-4
deepinfra/bf16bf1616k$0.070$0.140100.00%100.00%99.93%———
gemma-3-4b-it
google/gemma-3-4b-it
deepinfra/bf16bf16131k$0.050$0.100100.00%100.00%99.93%———
deepseek-v4-pro
deepseek/deepseek-v4-pro
deepinfra/fp8fp81.0M$1.30$2.60100.00%100.00%99.9%——是
glm-5.2
z-ai/glm-5.2
deepinfra/fp4fp41.0M$0.563$1.80100.00%100.00%99.9%——是
qwen3.5-9b
qwen/qwen3.5-9b
deepinfra/bf16bf16262k$0.100$0.150100.00%100.00%99.8%———
hermes-3-llama-3.1-405b
nousresearch/hermes-3-llama-3.1-405b
deepinfra/fp8fp8131k$1.00$1.00100.00%100.00%99.8%———
qwen3-next-80b-a3b-instruct
qwen/qwen3-next-80b-a3b-instruct
deepinfra/fp8fp8262k$0.090$1.10100.00%100.00%99.7%———
qwen3-vl-30b-a3b-instruct
qwen/qwen3-vl-30b-a3b-instruct
deepinfra/fp8fp8262k$0.150$0.600100.00%100.00%99.7%———
mistral-small-3.2-24b-instruct
mistralai/mistral-small-3.2-24b-instruct
deepinfra/fp8fp8128k$0.075$0.200100.00%100.00%99.7%———
gemma-3-27b-it
google/gemma-3-27b-it
deepinfra/fp8fp8131k$0.080$0.160100.00%100.00%99.7%———
kimi-k2.6
moonshotai/kimi-k2.6
deepinfra/fp4fp4262k$0.750$3.50100.00%100.00%99.6%——是
llama-3.1-70b-instruct
meta-llama/llama-3.1-70b-instruct
deepinfra/turbofp8131k$0.400$0.400100.00%100.00%99.6%———
kimi-k3
moonshotai/kimi-k3
deepinfra/mxfp4mxfp41.0M$2.85$14.3100.00%100.00%99.6%——是
kimi-k3
moonshotai/kimi-k3
deepinfra/mxfp4mxfp41.0M$2.85$14.3100.00%100.00%99.6%——是
glm-4.6
z-ai/glm-4.6
deepinfra/fp4fp4203k$0.500$2.00100.00%100.00%99.6%——是
nemotron-3.5-lightning
nvidia/nemotron-3.5-lightning
deepinfra/bf16bf16262k$0.060$0.160100.00%100.00%99.6%——是
nemotron-3.5-lightning
nvidia/nemotron-3.5-lightning
deepinfra/bf16bf16262k$0.060$0.160100.00%100.00%99.6%——是
qwen3.8-27b
qwen/qwen3.8-27b
deepinfra/bf16bf16262k$0.150$1.88100.00%100.00%99.6%——是
llama-4-scout
meta-llama/llama-4-scout
deepinfra/fp8fp8328k$0.100$0.300—100.00%99.5%———
deepseek-v3.2
deepseek/deepseek-v3.2
deepinfra/fp4fp4164k$0.260$0.380100.00%100.00%99.5%——是
gpt-oss-20b
openai/gpt-oss-20b
deepinfra/bf16bf16131k$0.030$0.140100.00%100.00%99.5%———
gpt-oss-20b
openai/gpt-oss-20b
deepinfra/bf16bf16131k$0.030$0.140100.00%100.00%99.5%———
qwen-2.5-72b-instruct
qwen/qwen-2.5-72b-instruct
deepinfra/fp8fp833k$0.360$0.400100.00%100.00%99.5%———
gemma-4-31b-it
google/gemma-4-31b-it
deepinfra/fp8fp8262k$0.200$0.400100.00%100.00%99.4%———
gemma-4-31b-it
google/gemma-4-31b-it
deepinfra/fp8fp8262k$0.200$0.400100.00%100.00%99.4%———
qwen3.5-35b-a3b
qwen/qwen3.5-35b-a3b
deepinfra/fp8fp8262k$0.140$1.00100.00%100.00%99.2%——是
deepseek-v4-flash-vision-exp
deepseek/deepseek-v4-flash-vision-exp
deepinfra/fp8fp81.0M$0.216$0.647100.00%100.00%99.0%——是
deepseek-v4.1-flash
deepseek/deepseek-v4.1-flash
deepinfra/fp8/flexfp81.0M$0.112$0.336100.00%100.00%98.9%——是
deepseek-v4.1-flash
deepseek/deepseek-v4.1-flash
deepinfra/fp8/flexfp81.0M$0.112$0.336100.00%100.00%98.9%——是
gemma-4-31b-it
google/gemma-4-31b-it
deepinfra/fp8/priorityfp8262k$0.300$0.600100.00%100.00%98.9%———
gemma-4-31b-it
google/gemma-4-31b-it
deepinfra/fp8/priorityfp8262k$0.300$0.600100.00%100.00%98.9%———
deepseek-v4.1-flash-1017
deepseek/deepseek-v4.1-flash-1017
deepinfra/fp8/flexfp81.0M$0.078$0.235100.00%100.00%98.9%——是
llama-guard-4-12b
meta-llama/llama-guard-4-12b
deepinfra/bf16bf16164k$0.180$0.180100.00%100.00%98.7%———
qwen3.5-397b-a17b
qwen/qwen3.5-397b-a17b
deepinfra/fp8fp8262k$0.450$3.00——98.6%——是
glm-5.3
z-ai/glm-5.3
deepinfra/fp4fp41.0M$0.563$2.50100.00%100.00%98.4%——是
glm-5.3
z-ai/glm-5.3
deepinfra/fp4fp41.0M$0.563$2.50100.00%100.00%98.4%——是
glm-4.7
z-ai/glm-4.7
deepinfra/fp4fp4203k$0.400$1.75100.00%100.00%98.4%——是
llama-guard-4-12b
meta-llama/llama-guard-4-12b
deepinfra/bf16/prioritybf16164k$0.270$0.270100.00%100.00%98.1%———
qwen3-vl-235b-a22b-instruct
qwen/qwen3-vl-235b-a22b-instruct
deepinfra/fp8fp8262k$0.200$0.880100.00%100.00%97.8%——是
qwen3.6-35b-a3b
qwen/qwen3.6-35b-a3b
deepinfra/fp8fp8262k$0.100$0.950100.00%100.00%97.3%——是
qwen3-235b-a22b-2507
qwen/qwen3-235b-a22b-2507
deepinfra/fp8fp8262k$0.090$0.550100.00%100.00%96.8%———
deepseek-chat
deepseek/deepseek-chat
deepinfra/fp4fp4164k$0.320$0.890100.00%100.00%96.3%———
ling-3.0-flash
inclusionai/ling-3.0-flash
deepinfra/bf16bf16131k$0.060$0.180100.00%96.7%96.1%——是
gpt-oss-120b
openai/gpt-oss-120b
deepinfra/bf16bf16131k$0.037$0.170100.00%100.00%94.1%———
gpt-oss-120b
openai/gpt-oss-120b
deepinfra/bf16bf16131k$0.037$0.170100.00%100.00%94.1%———
qwen3-coder
qwen/qwen3-coder
deepinfra/turbofp4262k$0.300$1.00100.00%85.7%93.9%——是
qwen3-32b
qwen/qwen3-32b
deepinfra/fp8fp841k$0.080$0.280100.00%66.7%92.4%———
llama-4-scout
meta-llama/llama-4-scout
deepinfra/fp8/priorityfp8328k$0.150$0.45080.0%93.3%90.6%———
gpt-oss-120b
openai/gpt-oss-120b
deepinfra/fp8fp8131k$0.200$0.950——70.2%———
gpt-oss-120b
openai/gpt-oss-120b
deepinfra/fp8fp8131k$0.200$0.950——70.2%———
llama-4-scout
meta-llama/llama-4-scout
deepinfra/fp8/flexfp8328k$0.080$0.240——59.1%———
deepseek-v4.1-flash
deepseek/deepseek-v4.1-flash
deepinfra/fp8/priorityfp81.0M$0.210$0.630100.00%100.00%13.5%——是
deepseek-v4.1-flash
deepseek/deepseek-v4.1-flash
deepinfra/fp8/priorityfp81.0M$0.210$0.630100.00%100.00%13.5%——是
phi-4
microsoft/phi-4
deepinfra/bf16/flexbf1616k$0.056$0.112——————
qwen3-next-80b-a3b-instruct
qwen/qwen3-next-80b-a3b-instruct
deepinfra/fp8/flexfp8262k$0.072$0.880——————
llama-guard-4-12b
meta-llama/llama-guard-4-12b
deepinfra/bf16/flexbf16164k$0.144$0.144——————

覆盖模型 · 69

元数据与端点观测同步自 OpenRouter 公开 API。GrokCode 不做路由转发;自建测速请用 API 实验室。

DeepInfra provider · models & uptime · GrokCode 倍率榜