Deep Infra
Inference hostDeep Infra runs open and licensed models on its own hardware and sells them by usage, often faster or cheaper than the maker.
Models88
- Cheapest for
- 24 models
- Serves
- 2 model types
Visits in September 2026502K−8.1%Unique visitors 170K
- Apr
- May
- Jun
- Jul
- Aug
- Sep
Models by type
Models on Deep Infra
| Model | Input | Output | vs official | Context |
|---|---|---|---|---|
| Kimi K3Moonshot AI | $2.85/1M | $14.25/1M from $8.28 at Kenari | +3% | — |
| Claude Opus 5.5Anthropic | $4/1M | $20/1M from $16 at APIMart | ±0% | — |
 GLM-5.3Zhipu AI | $0.563/1M | From$2.50/1M from $1.10 at NanoGPT | -43% | — |
 DeepSeek V4.1 FlashDeepSeek | $0.14/1M | From$0.42/1M from $0.03 at UnoRouter | +1300% | — |
| GLM-5.3-FlashZhipu AI | $0.15/1M | $0.50/1M from $0.20 at APIMart | ±0% | — |
 Qwen3.8 27BAlibaba | $0.15/1M | From$1.88/1M from $0.50 at EmpirioLabs AI | -37% | — |
| GLM-5.2Zhipu AICheapest | $0.563/1M | From$1.80/1M | -59% | — |
| Claude Sonnet 5.5Anthropic | $2/1M | $10/1M from $8 at APIMart | ±0% | — |
 MiMo-V2.6-ProXiaomi | $0.387/1M | $0.783/1M from $0.696 at CometAPI | +13% | — |
| DeepSeek V4 FlashDeepSeek | $0.09/1M | From$0.18/1M from $0.17 at Opper | +6% | — |
| DeepSeek V4 Flash Vision ExpDeepSeekCheapest | $0.098/1M | From$0.294/1M | -51% | — |
| Claude Opus 5Anthropic | $5/1M | $25/1M from $20 at APIMart | ±0% | — |
| DeepSeek V4 ProDeepSeek | $1.30/1M | From$2.60/1M from $0.418 at OpenRouter | +523% | — |
 Ling 3.1 FlashInclusionaiCheapest | $0.075/1M | $0.225/1M | ±0% | — |
| Claude Sonnet 4.6Anthropic | $3/1M | $15/1M from $12 at APIMart | ±0% | — |
| Claude Fable 5Anthropic | $10/1M | $50/1M from $40 at APIMart | ±0% | — |
| GPT OSS 120BOpenAI | $0.037/1M | From$0.17/1M from $0.14 at Requesty | +21% | — |
| MiMo-V2.6-FlashXiaomi | $0.091/1M | $0.182/1M from $0.10 at UnoRouter | +82% | — |
 MiniMax-M3MiniMax | $0.28/1M | From$1.10/1M from $0.90 at EmpirioLabs AI | -8% | — |
| Qwen3.8 MaxAlibaba | $1.65/1M | $4.95/1M from $4.11 at APIMart | ±0% | — |
| Claude Sonnet 5Anthropic | $3/1M | $15/1M from $6.90 at Kenari | +50% | — |
| Qwen3.8 FlashAlibaba | $0.113/1M | $0.382/1M from $0.309 at APIMart | ±0% | — |
 Gemini 3.1 ProGoogle | $2/1M | $12/1M | — | — |
| Qwen3.5 9BAlibaba | $0.10/1M | From$0.15/1M from $0.13 at EmpirioLabs AI | +15% | — |
| Gemini 2.5 ProGoogle | $1.25/1M | $10/1M from $8 at APIMart | ±0% | — |
| GPT OSS 20BOpenAI | $0.03/1M | From$0.14/1M from $0.09 at OpenRouter | +56% | — |
| Qwen3.6 35B-A3BAlibabaCheapest | $0.10/1M | From$0.95/1M | -58% | — |
| Gemini 2.5 FlashGoogle | $0.30/1M | $2.50/1M from $2 at APIMart | ±0% | — |
| Kimi K2.6Moonshot AI | $0.75/1M | From$3.50/1M from $2.40 at APIYI | -7% | — |
| DeepSeek V3DeepSeekCheapest | $0.32/1M | From$0.89/1M | ±0% | — |
 Llama 4 Scout 17B InstructMeta | $0.10/1M | From$0.30/1M from $0.20 at AIHubMix | +50% | — |
| Ling 3.0 Flash (free)Inclusionai | $0.06/1M | $0.18/1M from $0.063 at OpenRouter | +186% | — |
| GLM-4.7Zhipu AI | $0.40/1M | From$1.75/1M from $1.10 at AIHubMix | -20% | — |
| GLM-4.6Zhipu AI | $0.50/1M | From$2/1M from $1.10 at AIHubMix | -9% | — |
| Gemma 4Google | $0.20/1M | From$0.40/1M from $0.30 at FastRouter | +33% | — |
| Qwen3 30B A3BAlibaba | $0.12/1M | From$0.50/1M from $0.29 at Helicone | +72% | — |
 Nemotron 3.5 Lightning 30B A3BNVIDIACheapest | $0.06/1M | $0.16/1M | — | — |
| Llama-3.3-70B-InstructMetaCheapest | $0.10/1M | From$0.32/1M | — | — |
| Qwen3.6 27BAlibaba | $0.32/1M | From$3.20/1M from $2 at OpenRouter | -11% | — |
 Mistral NemoMistralCheapest | $0.029/1M | From$0.03/1M | ±0% | — |
| Gemma 3 4B ITGoogle | $0.05/1M | From$0.10/1M from $0.05 at OpenRouter | +100% | — |
| DeepSeek V4 Flash 0731DeepSeek | $0.06/1M | From$0.18/1M from $0.133 at OpenRouter | +35% | — |
| Qwen3 14BAlibaba | $0.12/1M | From$0.24/1M from $0.20 at Hugging Face | +20% | — |
| Qwen3.5 397B-A17BAlibaba | $0.45/1M | From$3/1M from $0.986 at AIHubMix | +191% | — |
| Qwen3 Coder 480B A35B InstructAlibaba | $0.30/1M | From$1/1M from $0.95 at Helicone | +5% | — |
| Qwen3.7 MaxAlibaba | $2.50/1M | $7.50/1M from $3.75 at DevPass (LLM Gateway) | +51% | — |
| Qwen3.5 35B-A3BAlibaba | $0.14/1M | From$1/1M from $0.451 at AIHubMix | -50% | — |
| Gemma 3 27BGoogleCheapest | $0.08/1M | From$0.16/1M | ±0% | — |
| Gemma 4 26B A4B ITGoogle | $0.07/1M | From$0.34/1M from $0.225 at OpenRouter | -43% | — |
| Qwen3 32BAlibaba | $0.08/1M | From$0.28/1M from $0.25 at Hugging Face | +12% | — |
Deep Infra at a glance
88 models in total: 86 text, 2 moderation. The cheapest is Mistral Nemo at $0.03 /1M. Llama 4 Scout 17B Instruct has the largest context window, 1.31M tokens. The most searched is Kimi K3, with about 658K Google searches a month. The newest is DeepSeek V4 Flash Vision Exp, released Oct 17, 2026.
Cheapest
- Mistral Nemo$0.03/1M
- SL3-8B-Lunaris-v1$0.05/1M
- Mistral Small 3$0.08/1M
- Gemma 3 4B IT$0.10/1M
- GPT OSS 20B$0.14/1M
Largest context
- Llama 4 Scout 17B Instruct1.31M
- Kimi K31.05M
- GLM-5.31.05M
- DeepSeek V4.1 Flash1.05M
- GLM-5.3-Flash1.05M
Most searched
- Kimi K3658K
- Claude Opus 5.5405K
- GLM-5.3276K
- DeepSeek V4.1 Flash138K
- GLM-5.3-Flash121K
Newest
- DeepSeek V4 Flash Vision ExpOct 17, 2026
- Nemotron 3.5 Super VL 120B A12BOct 15, 2026
- Ling 3.1 FlashOct 13, 2026
- Claude Sonnet 5.5Sep 28, 2026
- Claude Opus 5.5Sep 22, 2026