LLMRates.ai
DeepInfra logo

DeepInfra

モデルアグリゲーター

DeepInfra is a serverless inference marketplace running open models (Llama, Qwen, DeepSeek, GLM, Kimi and more) on managed GPUs, billed per token with no idle cost — among the cheapest hosted open-model APIs.

モデル
137
プロンプトキャッシュ
✓
バッチ割引
—
Tool Calling
✓
ウェブサイトを見る →公式料金を見る →
モデル137
モデル入力料金出力料金
anthropic/claude-fable-5¥1635¥8174
anthropic/claude-haiku-4-5¥163¥817
anthropic/claude-opus-4-7¥817¥4087
anthropic/claude-opus-4-8¥817¥4087
anthropic/claude-sonnet-4-6¥490¥2452
anthropic/claude-sonnet-5¥327¥1635
BAAI/bge-base-en-v1.5¥0.82—
BAAI/bge-en-icl¥2—
BAAI/bge-large-en-v1.5¥2—
BAAI/bge-m3¥2—
BAAI/bge-m3-multi¥2—
bosonai/HiggsAudioV2.5¥3269/100万文字—
ByteDance/Seed-1.8¥41¥327
ByteDance/Seed-2.0-code¥82¥490
ByteDance/Seed-2.0-mini¥16¥65
ByteDance/Seed-2.0-pro¥82¥490
ByteDance/Seedance-1.5-Pro¥196—
ByteDance/Seedance-2.0¥768—
canopylabs/orpheus-3b-0.1-ft¥1144/100万文字—
deepseek-ai/DeepSeek-R1-0528DeepSeek¥82¥351
deepseek-ai/DeepSeek-V3DeepSeek¥52¥145
deepseek-ai/DeepSeek-V3-0324DeepSeek¥39¥147
deepseek-ai/DeepSeek-V3.1DeepSeek¥41¥155
deepseek-ai/DeepSeek-V3.1-TerminusDeepSeek¥44¥155
deepseek-ai/DeepSeek-V3.2DeepSeek¥43¥62
deepseek-ai/DeepSeek-V4-FlashDeepSeek¥15¥29
deepseek-ai/DeepSeek-V4-ProDeepSeek¥213¥425
google/embeddinggemma-300mGemma¥0.33—
google/gemini-2.5-flash¥49¥409
google/gemini-2.5-pro¥204¥1635
google/gemini-3.1-flash-lite¥41¥245
google/gemini-3.1-pro¥327¥1962
google/gemini-3.5-flash¥245¥1471
google/gemini-3-pro-image¥19616—
google/gemma-3-12b-itGemma¥8¥25
google/gemma-3-27b-itGemma¥13¥26
google/gemma-3-4b-itGemma¥8¥16
google/gemma-4-26B-A4B-itGemma¥11¥56
google/gemma-4-31B-itGemma¥21¥62
google/gemma-4-31B-it-turboGemma¥20¥60
google/gemma-4-E4B-itGemma¥3¥16
Gryphe/MythoMax-L2-13b¥65¥65
hexgrad/Kokoro-82M¥101/100万文字—
intfloat/e5-base-v2¥0.82—
intfloat/e5-large-v2¥2—
intfloat/multilingual-e5-large¥2—
intfloat/multilingual-e5-large-instruct¥2—
inworld-ai/realtime-tts-1.5-max¥8174/100万文字—
inworld-ai/realtime-tts-1.5-mini¥4087/100万文字—
inworld-ai/realtime-tts-2¥5721/100万文字—
meta-llama/Llama-3.3-70B-Instruct-TurboLlama¥16¥52
meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8Llama¥33¥131
meta-llama/Llama-4-Scout-17B-16E-InstructLlama¥16¥49
meta-llama/Llama-Guard-4-12BLlama¥29¥29
meta-llama/Meta-Llama-3.1-70B-Instruct-TurboLlama¥65¥65
meta-llama/Meta-Llama-3.1-8B-Instruct-TurboLlama¥3¥7
microsoft/phi-4Phi¥11¥23
MiniMaxAI/MiniMax-M2.7MiniMax¥41¥163
MiniMaxAI/MiniMax-M2.7-TurboMiniMax¥62¥278
MiniMaxAI/MiniMax-M3MiniMax¥49¥196
mistralai/Mistral-Nemo-Instruct-2407Mistral¥3¥5
mistralai/Mistral-Small-24B-Instruct-2501Mistral¥8¥13
mistralai/Mistral-Small-3.2-24B-Instruct-2506Mistral¥12¥33
moonshotai/Kimi-K2.5Kimi¥74¥368
moonshotai/Kimi-K2.6Kimi¥123¥572
moonshotai/Kimi-K2.7-CodeKimi¥121¥572
NousResearch/Hermes-3-Llama-3.1-405BLlama¥163¥163
NousResearch/Hermes-3-Llama-3.1-70BLlama¥114¥114
nvidia/llama-nemotron-embed-vl-1b-v2Llama¥2—
nvidia/llama-nemotron-rerank-vl-1b-v2Llama¥2—
nvidia/Nemotron-3-Embed-1B-BF16Nemotron¥2—
nvidia/Nemotron-3-Embed-1B-NVFP4Nemotron¥2—
nvidia/Nemotron-3-Embed-8BNemotron¥6—
nvidia/Nemotron-3-Nano-30B-A3BNemotron¥8¥33
nvidia/Nemotron-Content-Safety-3.5Nemotron¥33¥33
nvidia/NVIDIA-Nemotron-3-Super-120B-A12BNemotron¥14¥65
nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55BNemotron¥82¥360
openai/gpt-oss-120bGPT-OSS¥6¥28
openai/gpt-oss-120b-TurboGPT-OSS¥25¥98
openai/gpt-oss-20bGPT-OSS¥5¥23
Qwen/Qwen2.5-72B-InstructQwen¥59¥65
Qwen/Qwen3-14BQwen¥20¥39
Qwen/Qwen3-235B-A22B-Instruct-2507Qwen¥15¥90
Qwen/Qwen3-235B-A22B-Thinking-2507Qwen¥38¥376
Qwen/Qwen3-30B-A3BQwen¥20¥82
Qwen/Qwen3-32BQwen¥13¥46
Qwen/Qwen3.5-122B-A10BQwen¥47¥392
Qwen/Qwen3.5-27BQwen¥43¥425
Qwen/Qwen3.5-35B-A3BQwen¥23¥163
Qwen/Qwen3.5-397B-A17BQwen¥74¥490
Qwen/Qwen3.5-9BQwen¥16¥25
Qwen/Qwen3.6-27BQwen¥52¥523
Qwen/Qwen3.6-35B-A3BQwen¥25¥155
Qwen/Qwen3.7-MaxQwen¥409¥1226
Qwen/Qwen3-Coder-480B-A35B-Instruct-TurboQwen¥49¥163
Qwen/Qwen3-Embedding-0.6BQwen¥2—
Qwen/Qwen3-Embedding-4BQwen¥3—
Qwen/Qwen3-Embedding-8BQwen¥2—
Qwen/Qwen3-MaxQwen¥196¥981
Qwen/Qwen3-Max-ThinkingQwen¥196¥981
Qwen/Qwen3-Next-80B-A3B-InstructQwen¥15¥180
Qwen/Qwen3-Reranker-0.6BQwen¥2—
Qwen/Qwen3-Reranker-4BQwen¥4—
Qwen/Qwen3-Reranker-8BQwen¥8—
Qwen/Qwen3-TTSQwen¥3269/100万文字—
Qwen/Qwen3-TTS-VoiceDesignQwen¥3269/100万文字—
Qwen/Qwen3-VL-235B-A22B-InstructQwen¥33¥144
Qwen/Qwen3-VL-30B-A3B-InstructQwen¥25¥98
ResembleAI/chatterbox-multilingual¥163/100万文字—
ResembleAI/chatterbox-turbo¥163/100万文字—
Sao10K/L3.1-70B-Euryale-v2.2¥139¥139
Sao10K/L3-8B-Lunaris-v1-Turbo¥7¥8
sentence-transformers/all-MiniLM-L12-v2¥0.82—
sentence-transformers/all-MiniLM-L6-v2¥0.82—
sentence-transformers/all-mpnet-base-v2¥0.82—
sentence-transformers/clip-ViT-B-32¥0.82—
sentence-transformers/clip-ViT-B-32-multilingual-v1¥0.82—
sentence-transformers/multi-qa-mpnet-base-dot-v1¥0.82—
sentence-transformers/paraphrase-MiniLM-L6-v2¥0.82—
sesame/csm-1b¥1144/100万文字—
shibing624/text2vec-base-chinese¥0.82—
stepfun-ai/Step-3.7-Flash¥33¥188
tencent/Hy3¥23¥95
thenlper/gte-base¥0.82—
thenlper/gte-large¥2—
thinkingmachines/Inkling¥163¥662
XiaomiMiMo/MiMo-V2.5¥65¥327
XiaomiMiMo/MiMo-V2.5-Pro¥163¥490
XiaomiMiMo/MiMo-V2.5-ttsFree/100万文字—
XiaomiMiMo/MiMo-V2.5-tts-voicecloneFree/100万文字—
XiaomiMiMo/MiMo-V2.5-tts-voicedesignFree/100万文字—
zai-org/GLM-4.6GLM¥82¥327
zai-org/GLM-4.7GLM¥65¥286
zai-org/GLM-4.7-FlashGLM¥10¥65
zai-org/GLM-5GLM¥98¥340
zai-org/GLM-5.1GLM¥172¥572
zai-org/GLM-5.2GLM¥152¥490