Deep Infra is a cloud platform that resells models with its own infrastructure, billing and regional hosting. Its prices tend to track the first-party rate, with the platform's own quotas and service-level terms attached.
It lists 80 models on this page. Coverage spans text and chat (79) and classification and safety (1). The lowest-priced model hosted here is Gemma 4 E4B IT.
Deep Infra is the lowest paid channel for 11 models that have at least two competing paid hosts, which is where its pricing genuinely leads rather than being the only option.
| Cheapest here? | |||||
|---|---|---|---|---|---|
| Gemma 4 E4B ITgoogle/gemma-4-E4B-itLowest anywhere | $0.02 | $0.10 | — | 131,072 | Lowest anywhere |
| GPT OSS 20Bopenai/gpt-oss-20b3.2× pricier | $0.03 | $0.14 | $0.015 | 131,072 | 3.2× pricier |
| Mistral Small 3mistralai/Mistral-Small-24B-Instruct-2501Lowest anywhere | $0.05 | $0.08 | — | 32,768 | Lowest anywhere |
| Gemma 3 4B ITgoogle/gemma-3-4b-it1.5× pricier | $0.05 | $0.10 | — | 131,072 | 1.5× pricier |
| GPT OSS 120Bopenai/gpt-oss-120b1.2× pricier | $0.037 | $0.17 | $0.019 | 131,072 | 1.2× pricier |
| Gemma 3 12B ITgoogle/gemma-3-12b-it1.3× pricier | $0.05 | $0.15 | — | 131,072 | 1.3× pricier |
| Nemotron 3 Nano 30B A3Bnvidia/Nemotron-3-Nano-30B-A3BLowest anywhere | $0.05 | $0.20 | $0.025 | 262,144 | Lowest anywhere |
| DeepSeek V4 Flash 0731deepseek-ai/DeepSeek-V4-Flash-07311.8× pricier | $0.06 | $0.18 | $0.015 | 1,048,576 | 1.8× pricier |
| Gemma 3 27B ITgoogle/gemma-3-27b-it1.3× pricier | $0.08 | $0.16 | — | 131,072 | 1.3× pricier |
| DeepSeek V4 Flashdeepseek-ai/DeepSeek-V4-Flash1.4× pricier | $0.09 | $0.18 | $0.018 | 1,048,576 | 1.4× pricier |
| Qwen3.5 9BQwen/Qwen3.5-9B1.1× pricier | $0.10 | $0.15 | — | 262,144 | 1.1× pricier |
| Qwen3 32BQwen/Qwen3-32BLowest anywhere | $0.08 | $0.28 | — | 40,960 | Lowest anywhere |
| Gemma 4 26B A4B ITgoogle/gemma-4-26B-A4B-it2.7× pricier | $0.07 | $0.34 | — | 262,144 | 2.7× pricier |
| GLM-4.7-Flashzai-org/GLM-4.7-FlashLowest anywhere | $0.06 | $0.40 | $0.01 | 202,752 | Lowest anywhere |
| Llama 4 Scout 17B Instructmeta-llama/Llama-4-Scout-17B-16E-Instruct1.1× pricier | $0.10 | $0.30 | — | 327,680 | 1.1× pricier |
| Llama-3.3-70B-Instructmeta-llama/Llama-3.3-70B-Instruct-Turbo1.6× pricier | $0.10 | $0.32 | — | 131,072 | 1.6× pricier |
| MiMo-V2.6-FlashXiaomiMiMo/MiMo-V2.6-Flash1.3× pricier | $0.14 | $0.28 | $0.0028 | 1,048,576 | 1.3× pricier |
| MiMo-V2.5XiaomiMiMo/MiMo-V2.51.8× pricier | $0.14 | $0.28 | $0.0028 | 262,144 | 1.8× pricier |
| Seed 2.0 MiniByteDance/Seed-2.0-mini1.9× pricier | $0.10 | $0.40 | $0.02 | 256,000 | 1.9× pricier |
| Llama Guard 4 12BClassify / Safetymeta-llama/Llama-Guard-4-12B1.3× pricier | $0.18 | $0.18 | — | 163,840 | 1.3× pricier |
| Qwen3.8 FlashQwen/Qwen3.8-Flash1.0× pricier | $0.11 | $0.38 | $0.014 | 1,000,000 | 1.0× pricier |
| Qwen3 235B-A22B Instruct 2507Qwen/Qwen3-235B-A22B-Instruct-25072.2× pricier | $0.09 | $0.55 | — | 262,144 | 2.2× pricier |
| Qwen3 30B A3BQwen/Qwen3-30B-A3B1.9× pricier | $0.12 | $0.50 | — | 40,960 | 1.9× pricier |
| Hy3tencent/Hy32.0× pricier | $0.13 | $0.53 | $0.033 | 262,144 | 2.0× pricier |
| GLM-5.3-Flashzai-org/GLM-5.3-Flash4.8× pricier | $0.15 | $0.50 | $0.03 | 1,048,576 | 4.8× pricier |
| Gemma 4 31B ITgoogle/gemma-4-31B-it1.8× pricier | $0.20 | $0.40 | — | 262,144 | 1.8× pricier |
| Qwen3 VL 30B A3B InstructQwen/Qwen3-VL-30B-A3B-Instruct1.4× pricier | $0.15 | $0.60 | — | 262,144 | 1.4× pricier |
| DeepSeek V3.2deepseek-ai/DeepSeek-V3.21.2× pricier | $0.26 | $0.38 | $0.13 | 163,840 | 1.2× pricier |
| DeepSeek V4.1 Flash 1017deepseek-ai/DeepSeek-V4.1-Flash-10175.7× pricier | $0.20 | $0.60 | $0.006 | 1,048,576 | 5.7× pricier |
| DeepSeek V4.1 Flashdeepseek-ai/DeepSeek-V4.1-Flash5.7× pricier | $0.20 | $0.60 | $0.006 | 1,048,576 | 5.7× pricier |
| Qwen3.6 35B-A3BQwen/Qwen3.6-35B-A3B2.9× pricier | $0.10 | $0.95 | $0.10 | 262,144 | 2.9× pricier |
| Qwen3-Next 80B-A3B InstructQwen/Qwen3-Next-80B-A3B-Instruct1.4× pricier | $0.09 | $1.10 | — | 262,144 | 1.4× pricier |
| Llama 4 Maverick 17B Instructmeta-llama/Llama-4-Maverick-17B-128E-Instruct-FP81.4× pricier | $0.20 | $0.80 | — | 1,048,576 | 1.4× pricier |
| Nemotron 3 Nano Omni 30B A3B Reasoningnvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning2.5× pricier | $0.20 | $0.80 | — | 262,144 | 2.5× pricier |
| Qwen3.5 35B-A3BQwen/Qwen3.5-35B-A3B2.3× pricier | $0.14 | $1.00 | $0.05 | 262,144 | 2.3× pricier |
| Qwen3 VL 235B A22B InstructQwen/Qwen3-VL-235B-A22B-InstructLowest anywhere | $0.20 | $0.88 | $0.11 | 262,144 | Lowest anywhere |
| MiniMax-M2.5MiniMaxAI/MiniMax-M2.51.0× pricier | $0.15 | $1.15 | $0.03 | 196,608 | 1.0× pricier |
| Llama 3.3 Nemotron Super 49B v1.5nvidia/Llama-3.3-Nemotron-Super-49B-v1.5Only channel | $0.40 | $0.40 | — | 131,072 | Only channel |
| DeepSeek V3 0324deepseek-ai/DeepSeek-V3-03241.2× pricier | $0.24 | $0.90 | $0.14 | 163,840 | 1.2× pricier |
| DeepSeek-V3.1deepseek-ai/DeepSeek-V3.11.3× pricier | $0.25 | $0.95 | $0.13 | 163,840 | 1.3× pricier |
| MiniMax-M2.7MiniMaxAI/MiniMax-M2.72.2× pricier | $0.25 | $1.00 | $0.05 | 196,608 | 2.2× pricier |
| Step 3.7 Flashstepfun-ai/Step-3.7-FlashLowest anywhere | $0.20 | $1.15 | $0.04 | 262,144 | Lowest anywhere |
| DeepSeek-V3deepseek-ai/DeepSeek-V3Lowest anywhere | $0.32 | $0.89 | — | 163,840 | Lowest anywhere |
| Qwen3-Coder 480B-A35B InstructQwen/Qwen3-Coder-480B-A35B-Instruct-Turbo2.1× pricier | $0.30 | $1.00 | $0.10 | 262,144 | 2.1× pricier |
| Inkling 2 Smallthinkingmachines/Inkling-2-Small2.0× pricier | $0.30 | $1.00 | $0.03 | 1,048,576 | 2.0× pricier |
| MiniMax-M3MiniMaxAI/MiniMax-M31.3× pricier | $0.28 | $1.10 | $0.056 | 524,288 | 1.3× pricier |
| MiMo-V2.6-ProXiaomiMiMo/MiMo-V2.6-Pro1.5× pricier | $0.43 | $0.87 | $0.0036 | 1,048,576 | 1.5× pricier |
| Inkling Smallthinkingmachines/Inkling-Small1.2× pricier | $0.45 | $1.20 | $0.10 | 524,288 | 1.2× pricier |
| DeepSeek V4 Flash Vision Expdeepseek-ai/DeepSeek-V4-Flash-Vision-Exp3.8× pricier | $0.44 | $1.32 | $0.014 | 1,048,576 | 3.8× pricier |
| GLM-4.7zai-org/GLM-4.71.5× pricier | $0.40 | $1.75 | $0.08 | 202,752 | 1.5× pricier |
| Qwen3.8 27BQwen/Qwen3.8-27B6.8× pricier | $0.20 | $2.50 | $0.05 | 262,144 | 6.8× pricier |
| Qwen3.5 122B-A10BQwen/Qwen3.5-122B-A10B2.6× pricier | $0.29 | $2.40 | — | 262,144 | 2.6× pricier |
| Qwen3.5 27BQwen/Qwen3.5-27B3.6× pricier | $0.26 | $2.60 | — | 262,144 | 3.6× pricier |
| GLM-4.6zai-org/GLM-4.61.8× pricier | $0.50 | $2.00 | $0.10 | 202,752 | 1.8× pricier |
| Kimi K2.5moonshotai/Kimi-K2.5Lowest anywhere | $0.45 | $2.25 | $0.07 | 262,144 | Lowest anywhere |
| DeepSeek R1 0528deepseek-ai/DeepSeek-R1-05282.6× pricier | $0.50 | $2.15 | $0.35 | 163,840 | 2.6× pricier |
| GLM-5zai-org/GLM-51.3× pricier | $0.60 | $2.08 | $0.12 | 202,752 | 1.3× pricier |
| Qwen3.6 27BQwen/Qwen3.6-27B2.9× pricier | $0.32 | $3.20 | — | 262,144 | 2.9× pricier |
| Qwen3.5 397B-A17BQwen/Qwen3.5-397B-A17B2.9× pricier | $0.45 | $3.00 | $0.22 | 262,144 | 2.9× pricier |
| Seed 2.0 CodeByteDance/Seed-2.0-code1.3× pricier | $0.50 | $3.00 | $0.10 | 256,000 | 1.3× pricier |
| Seed 2.0 ProByteDance/Seed-2.0-pro1.3× pricier | $0.50 | $3.00 | $0.10 | 256,000 | 1.3× pricier |
| GLM-5.2zai-org/GLM-5.21.8× pricier | $0.75 | $2.40 | $0.14 | 1,048,576 | 1.8× pricier |
| Hy4 previewtencent/Hy4-preview1.2× pricier | $0.83 | $2.50 | $0.042 | 1,048,576 | 1.2× pricier |
| Kimi K2.7 Codemoonshotai/Kimi-K2.7-Code2.8× pricier | $0.68 | $3.40 | $0.14 | 262,144 | 2.8× pricier |
| Step 5 Previewstepfun-ai/Step-5-Preview2.2× pricier | $1.00 | $2.70 | $0.10 | 1,048,576 | 2.2× pricier |
| Kimi K2.6moonshotai/Kimi-K2.64.6× pricier | $0.75 | $3.50 | $0.15 | 262,144 | 4.6× pricier |
| MiMo-V2.5-ProXiaomiMiMo/MiMo-V2.5-Pro6.0× pricier | $1.00 | $3.00 | $0.20 | 1,048,576 | 6.0× pricier |
| DeepSeek V4 Prodeepseek-ai/DeepSeek-V4-Pro3.7× pricier | $1.30 | $2.60 | $0.10 | 1,048,576 | 3.7× pricier |
| DeepSeek V4 Pro 0813deepseek-ai/DeepSeek-V4-Pro-08131.8× pricier | $1.30 | $2.60 | $0.10 | 1,048,576 | 1.8× pricier |
| GLM-5.1zai-org/GLM-5.11.5× pricier | $1.05 | $3.50 | $0.21 | 202,752 | 1.5× pricier |
| GLM-5.3zai-org/GLM-5.34.9× pricier | $0.90 | $4.00 | $0.20 | 1,048,576 | 4.9× pricier |
| Inklingthinkingmachines/InklingLowest anywhere | $0.95 | $4.05 | $0.16 | 524,288 | Lowest anywhere |
| Qwen4 MaxQwen/Qwen4-MaxLowest anywhere | $1.41 | $4.24 | $0.071 | 1,000,000 | Lowest anywhere |
| Qwen3 MaxQwen/Qwen3-Max3.8× pricier | $1.20 | $6.00 | $0.24 | 256,000 | 3.8× pricier |
| Qwen3.8 MaxQwen/Qwen3.8-Max1.1× pricier | $1.65 | $4.95 | $0.21 | 256,000 | 1.1× pricier |
| Qwen3.8 2.4T A95BQwen/Qwen3.8-2.4T-A95B1.1× pricier | $2.00 | $6.00 | $0.20 | 262,144 | 1.1× pricier |
| Qwen3.7 MaxQwen/Qwen3.7-Max3.0× pricier | $2.50 | $7.50 | $0.50 | 256,000 | 3.0× pricier |
| Kimi K3.1 Promoonshotai/Kimi-K3.1-Pro1.7× pricier | $2.00 | $10.00 | $0.20 | 1,048,576 | 1.7× pricier |
| Kimi K3moonshotai/Kimi-K33.2× pricier | $2.85 | $14.25 | $0.29 | 1,048,576 | 3.2× pricier |
| Claude Fable 5anthropic/claude-fable-52.9× pricier | $10.00 | $50.00 | — | 1,000,000 | 2.9× pricier |