Novita AI models & pricing
Novita AI hosts 135 models (135 with public pricing) covering 3 modalities. Pay-as-you-go GPU inference for Llama, Qwen, FLUX, and image/video models. Cheapest input starts at $0.0100/M tokens; the most premium goes up to $4.00/M. Use Future AGI's Agent Command Center to route any Novita AI model with cost-optimized fallback and unified observability.
135
chat 130
| Model ↕ | Input / 1M ↕ | Output / 1M ↕ | Blended ↕ | Context ↕ | Caps |
|---|---|---|---|---|---|
| Llama 3.1 8B Instruct | $0.0200/M | $0.0500/M | $0.0275/M | 16,384 | |
| Llama 3.2 1B Instruct | $0.0200/M | $0.0200/M | $0.0200/M | 131,000 | |
| Paddlepaddle Paddleocr VL | $0.0200/M | $0.0200/M | $0.0200/M | 16,384 | vision |
| DeepSeek OCR | $0.0300/M | $0.0300/M | $0.0300/M | 8,192 | vision |
| DeepSeek OCR 2 | $0.0300/M | $0.0300/M | $0.0300/M | 8,192 | vision |
| Llama 3.2 3B Instruct | $0.0300/M | $0.0500/M | $0.0350/M | 32,768 | tools |
| Qwen3 4B FP8 | $0.0300/M | $0.0300/M | $0.0300/M | 128,000 | tools · reasoning |
| Qwen3 8B FP8 | $0.0350/M | $0.138/M | $0.0608/M | 128,000 | reasoning |
| Zai Org Autoglm Phone 9B Multilingual | $0.0350/M | $0.138/M | $0.0608/M | 65,536 | vision |
| Llama 3 8B Instruct | $0.0400/M | $0.0400/M | $0.0400/M | 8,192 | |
| Mistral Nemo | $0.0400/M | $0.170/M | $0.0725/M | 60,288 | |
| gpt-oss 20B | $0.0400/M | $0.150/M | $0.0675/M | 131,072 | reasoning |
| Gemma 3 12B IT | $0.0500/M | $0.1000/M | $0.0625/M | 131,072 | vision |
| Nvidia Nemotron 3 nano 30B A3B | $0.0500/M | $0.200/M | $0.0875/M | 262,144 | tools · reasoning |
| gpt-oss 120B | $0.0500/M | $0.250/M | $0.100/M | 131,072 | tools · reasoning |
| Sao10k L3 8B Lunaris | $0.0500/M | $0.0500/M | $0.0500/M | 8,192 | |
| Sao10k L3 8B Stheno v3.2 | $0.0500/M | $0.0500/M | $0.0500/M | 8,192 | tools |
| DeepSeek R1 0528 Qwen3 8B | $0.0600/M | $0.0900/M | $0.0675/M | 128,000 | reasoning |
| Inclusionai Ling 3.0 Flash | $0.0600/M | $0.180/M | $0.0900/M | 262,144 | tools · reasoning · cache |
| Inclusionai Ling 3.0 Flash Fast | $0.0600/M | $0.180/M | $0.0900/M | 262,144 | tools · reasoning · cache |
| Baichuan Baichuan M2 32B | $0.0700/M | $0.0700/M | $0.0700/M | 131,072 | |
| Baidu Ernie 4.5 21B A3B | $0.0700/M | $0.280/M | $0.123/M | 120,000 | tools |
| Baidu Ernie 4.5 21B A3B Thinking | $0.0700/M | $0.280/M | $0.123/M | 131,072 | reasoning |
| Qwen2.5 7B Instruct | $0.0700/M | $0.0700/M | $0.0700/M | 32,000 | tools |
| Qwen3 Coder 30B A3B Instruct | $0.0700/M | $0.270/M | $0.120/M | 160,000 | tools |
| GLM 4.7 Flash | $0.0700/M | $0.400/M | $0.153/M | 200,000 | tools · reasoning · cache |
| Qwen3 VL 8B Instruct | $0.0800/M | $0.500/M | $0.185/M | 131,072 | tools · vision |
| Gryphe Mythomax L2 13B | $0.0900/M | $0.0900/M | $0.0900/M | 4,096 | |
| Qwen3 235B A22B Instruct 2507 | $0.0900/M | $0.580/M | $0.213/M | 131,072 | tools |
| Qwen3 30B A3B FP8 | $0.0900/M | $0.450/M | $0.180/M | 40,960 | tools · reasoning |
| Qwen3 32B FP8 | $0.1000/M | $0.450/M | $0.188/M | 40,960 | reasoning |
| Mimo V2 Flash | $0.110/M | $0.330/M | $0.165/M | 262,144 | tools · reasoning |
| Gemma 3 27B IT | $0.119/M | $0.200/M | $0.139/M | 98,304 | vision |
| Gemma 4 26B A4B IT | $0.130/M | $0.400/M | $0.198/M | 262,144 | tools · vision · reasoning |
| GLM 4.5 Air | $0.130/M | $0.850/M | $0.310/M | 131,072 | tools · reasoning · cache |
| Llama 3.3 70B Instruct | $0.135/M | $0.400/M | $0.201/M | 12,288 | tools |
| Baidu Ernie 4.5 VL 28B A3B | $0.140/M | $0.560/M | $0.245/M | 30,000 | tools · vision · reasoning |
| DeepSeek V4 Flash | $0.140/M | $0.280/M | $0.175/M | 1,048,576 | tools · reasoning · cache |
| Gemma 4 31B IT | $0.140/M | $0.400/M | $0.205/M | 262,144 | tools · vision · reasoning |
| Nousresearch Hermes 2 Pro Llama 3 8B | $0.140/M | $0.140/M | $0.140/M | 8,192 | |
| Tencent Hy3 | $0.140/M | $0.580/M | $0.250/M | 262,144 | tools · reasoning · cache |
| DeepSeek R1 Distill Qwen 14B | $0.150/M | $0.150/M | $0.150/M | 32,768 | reasoning |
| Qwen3 Next 80B A3B Instruct | $0.150/M | $1.50/M | $0.488/M | 131,072 | tools · reasoning |
| Qwen3 Next 80B A3B Thinking | $0.150/M | $1.50/M | $0.488/M | 131,072 | tools · reasoning |
| Mimo v2.5 | $0.168/M | $0.336/M | $0.210/M | 1,048,576 | tools · vision · reasoning · cache |
| Llama 4 Scout 17B 16e Instruct | $0.180/M | $0.590/M | $0.283/M | 131,072 | vision |
| Qwen3 235B A22B FP8 | $0.200/M | $0.800/M | $0.350/M | 40,960 | reasoning |
| Qwen3 Coder Next | $0.200/M | $1.50/M | $0.525/M | 262,144 | tools |
| Qwen3 VL 30B A3B Instruct | $0.200/M | $0.700/M | $0.325/M | 131,072 | tools · vision |
| Qwen3 VL 30B A3B Thinking | $0.200/M | $1.00/M | $0.400/M | 131,072 | tools · vision |
| Skywork R1v4 Lite | $0.200/M | $0.600/M | $0.300/M | 262,144 | vision |
| Stepfun Step 3.7 Flash | $0.200/M | $1.15/M | $0.438/M | 262,144 | tools · vision · reasoning · cache |
| Qwen3.6 35B A3B | $0.248/M | $1.48/M | $0.557/M | 262,144 | tools · vision · reasoning |
| Qwen Mt Plus | $0.250/M | $0.750/M | $0.375/M | 16,384 | |
| Qwen3.5 35B A3B | $0.250/M | $2.00/M | $0.688/M | 262,144 | tools · vision · reasoning |
| Qwen3 Omni 30B A3B Instruct | $0.250/M | $0.970/M | $0.430/M | 65,536 | tools · vision · audio |
| Qwen3 Omni 30B A3B Thinking | $0.250/M | $0.970/M | $0.430/M | 65,536 | tools · vision · reasoning · audio |
| DeepSeek v3.2 | $0.269/M | $0.400/M | $0.302/M | 163,840 | tools · reasoning |
| DeepSeek v3.0324 | $0.270/M | $1.12/M | $0.483/M | 163,840 | tools |
| DeepSeek v3.1 | $0.270/M | $1.00/M | $0.453/M | 131,072 | tools · reasoning |
| DeepSeek V3.1 Terminus | $0.270/M | $1.00/M | $0.453/M | 131,072 | tools · reasoning |
| DeepSeek V3.2 exp | $0.270/M | $0.410/M | $0.305/M | 163,840 | tools · reasoning |
| Llama 4 Maverick 17B 128e Instruct FP8 | $0.270/M | $0.850/M | $0.415/M | 1,048,576 | vision |
| Baidu Cobuddy | $0.280/M | $1.13/M | $0.493/M | 131,072 | tools · reasoning · cache |
| Baidu Ernie 4.5 300B A47B Paddle | $0.280/M | $1.10/M | $0.485/M | 123,000 | |
| DeepSeek R1 Distill Qwen 32B | $0.300/M | $0.300/M | $0.300/M | 64,000 | reasoning |
| Kwaipilot Kat Coder Pro | $0.300/M | $1.20/M | $0.525/M | 256,000 | tools |
| MiniMax M2 | $0.300/M | $1.20/M | $0.525/M | 204,800 | tools · reasoning |
| MiniMax M2.1 | $0.300/M | $1.20/M | $0.525/M | 204,800 | tools |
| MiniMax M2.5 | $0.300/M | $1.20/M | $0.525/M | 204,800 | tools · reasoning · cache |
| MiniMax M2.7 | $0.300/M | $1.20/M | $0.525/M | 204,800 | tools · reasoning · cache |
| MiniMax M3 | $0.300/M | $1.20/M | $0.525/M | 1,000,000 | tools · vision · reasoning · cache |
| Qwen3 235B A22B Thinking 2507 | $0.300/M | $3.00/M | $0.975/M | 131,072 | tools · reasoning |
| Qwen3.5 27B | $0.300/M | $2.40/M | $0.825/M | 262,144 | tools · vision · reasoning |
| Qwen3 VL 235B A22B Instruct | $0.300/M | $1.50/M | $0.600/M | 131,072 | tools · vision |
| GLM 4.6v | $0.300/M | $0.900/M | $0.450/M | 131,072 | tools · vision · reasoning |
| Qwen 2.5 72B Instruct | $0.380/M | $0.400/M | $0.385/M | 32,000 | tools |
| Qwen3 Coder 480B A35B Instruct | $0.380/M | $1.55/M | $0.673/M | 262,144 | tools |
| Baidu Ernie 4.5 VL 28B A3B Thinking | $0.390/M | $0.390/M | $0.390/M | 131,072 | tools · vision · reasoning |
| DeepSeek V3 Turbo | $0.400/M | $1.30/M | $0.625/M | 64,000 | tools |
| Qwen3.5 122B A10B | $0.400/M | $3.20/M | $1.10/M | 262,144 | tools · vision · reasoning |
| Baidu Ernie 4.5 VL 424B A47B | $0.420/M | $1.25/M | $0.628/M | 123,000 | vision · reasoning |
| DeepSeek V4 Flash 0731 | $0.440/M | $1.32/M | $0.660/M | 1,048,576 | tools · reasoning · cache |
| DeepSeek V4 Flash Vision exp | $0.440/M | $1.32/M | $0.660/M | 1,048,576 | tools · vision · reasoning · cache |
| Mindai Macaron V1 Tall | $0.450/M | $2.60/M | $0.988/M | 262,144 | tools · reasoning · cache |
| Llama 3 70B Instruct | $0.510/M | $0.740/M | $0.568/M | 8,192 | |
| Mimo V2.5 Pro | $0.522/M | $1.04/M | $0.653/M | 1,048,576 | tools · reasoning · cache |
| MiniMax M1 80K | $0.550/M | $2.20/M | $0.963/M | 1,000,000 | tools · reasoning |
| GLM 4 32B 0414 | $0.550/M | $1.66/M | $0.828/M | 32,000 | tools |
| GLM 4.6 | $0.550/M | $2.20/M | $0.963/M | 204,800 | tools · reasoning |
| Kimi K2 Instruct | $0.570/M | $2.30/M | $1.00/M | 131,072 | tools |
| MiniMax M2.5 Highspeed | $0.600/M | $2.40/M | $1.05/M | 204,800 | tools · reasoning · cache |
| MiniMax M2.7 Highspeed | $0.600/M | $2.40/M | $1.05/M | 204,800 | tools · reasoning · cache |
| Kimi K2 0905 | $0.600/M | $2.50/M | $1.08/M | 262,144 | tools |
| Kimi K2.5 | $0.600/M | $3.00/M | $1.20/M | 262,144 | tools · vision · reasoning · cache |
| Kimi K2 Thinking | $0.600/M | $2.50/M | $1.08/M | 262,144 | tools · reasoning · cache |
| Qwen3.5 397B A17B | $0.600/M | $3.60/M | $1.35/M | 262,144 | tools · vision · reasoning |
| Qwen3.6 27B | $0.600/M | $3.60/M | $1.35/M | 262,144 | tools · vision · reasoning |
| GLM 4.5 | $0.600/M | $2.20/M | $1.00/M | 131,072 | tools · reasoning |
| GLM 4.5v | $0.600/M | $1.80/M | $0.900/M | 65,536 | tools · vision · reasoning |
| GLM 4.7 | $0.600/M | $2.20/M | $1.00/M | 204,800 | tools · reasoning |
| GLM 4.7 H | $0.600/M | $2.20/M | $1.00/M | 204,800 | tools · reasoning · cache |
| Microsoft Wizardlm 2.8x22b | $0.620/M | $0.620/M | $0.620/M | 65,535 | |
| DeepSeek Prover V2 671B | $0.700/M | $2.50/M | $1.15/M | 160,000 | |
| DeepSeek R1 0528 | $0.700/M | $2.50/M | $1.15/M | 163,840 | tools · reasoning |
| DeepSeek R1 Turbo | $0.700/M | $2.50/M | $1.15/M | 64,000 | tools · reasoning |
| DeepSeek R1 Distill Llama 70B | $0.800/M | $0.800/M | $0.800/M | 8,192 | reasoning |
| Kimi K2.6 | $0.800/M | $3.40/M | $1.45/M | 262,144 | tools · vision · reasoning · cache |
| Qwen2.5 VL 72B Instruct | $0.800/M | $0.800/M | $0.800/M | 32,768 | vision |
| DeepSeek V3 | $0.890/M | $0.890/M | $0.890/M | 64,000 | tools |
| DeepSeek V3 Community | $0.890/M | $0.890/M | $0.890/M | 64,000 | tools |
| Kimi K2.7 Code | $0.950/M | $4.00/M | $1.71/M | 262,144 | tools · vision · reasoning · cache |
| Qwen3 VL 235B A22B Thinking | $0.980/M | $3.95/M | $1.72/M | 131,072 | tools · vision · reasoning |
| GLM 5 | $1.00/M | $3.20/M | $1.55/M | 202,800 | tools · reasoning · cache |
| GLM 5 Turbo | $1.20/M | $4.00/M | $1.90/M | 202,800 | tools · reasoning · cache |
| GLM 5v Turbo | $1.20/M | $4.00/M | $1.90/M | 204,800 | tools · vision · reasoning · cache |
| Qwen3.7 Max | $1.25/M | $3.75/M | $1.88/M | 1,000,000 | tools · reasoning · cache |
| DeepSeek V4 Pro 0813 | $1.32/M | $3.96/M | $1.98/M | 1,048,576 | tools · reasoning · cache |
| GLM 5.1 | $1.38/M | $4.40/M | $2.14/M | 204,800 | tools · reasoning · cache |
| GLM 5.2 | $1.40/M | $4.40/M | $2.15/M | 1,048,576 | tools · reasoning · cache |
| GLM 5.3 | $1.40/M | $4.40/M | $2.15/M | 1,048,576 | tools · reasoning · cache |
| Sao10k L3 70B Euryale v2.1 | $1.48/M | $1.48/M | $1.48/M | 8,192 | tools |
| Sao10k L31 70B Euryale v2.2 | $1.48/M | $1.48/M | $1.48/M | 8,192 | tools |
| Mindai Macaron V1 Venti | $1.50/M | $4.50/M | $2.25/M | 1,048,576 | tools · reasoning · cache |
| DeepSeek V4 Pro | $1.60/M | $3.20/M | $2.00/M | 1,048,576 | tools · reasoning · cache |
| Qwen3.8 Max | $2.00/M | $6.00/M | $3.00/M | 1,000,000 | tools · vision · reasoning · cache |
| Qwen3 Max | $2.11/M | $8.45/M | $3.70/M | 262,144 | tools |
| Kimi K3 | $3.00/M | $15.00/M | $6.00/M | 1,048,576 | tools · vision · reasoning · cache |
| DeepSeek R1 | $4.00/M | $4.00/M | $4.00/M | 64,000 | tools · reasoning |
| DeepSeek R1 Community | $4.00/M | $4.00/M | $4.00/M | 64,000 | tools · reasoning |
embedding 3
| Model ↕ | Input / 1M ↕ | Output / 1M ↕ | Blended ↕ | Context ↕ | Caps |
|---|---|---|---|---|---|
| Baai BGE M3 | $0.0100/M | $0.0100/M | $0.0100/M | 8,192 | |
| Qwen3 Embedding 0 6B | $0.0700/M | — | — | 32,768 | |
| Qwen3 Embedding 8B | $0.0700/M | — | — | 32,768 |
rerank 2
| Model ↕ | Input / 1M ↕ | Output / 1M ↕ | Blended ↕ | Context ↕ | Caps |
|---|---|---|---|---|---|
| Baai BGE Reranker V2 M3 | $0.0100/M | $0.0100/M | $0.0100/M | 8,000 | |
| Qwen3 Reranker 8B | $0.0500/M | $0.0500/M | $0.0500/M | 32,768 |
FAQ
How many Novita AI models are there?
135 Novita AI models are listed across 3 modalities on this page. 135 have public per-token pricing.
How is Novita AI pricing verified?
Pricing is aggregated from BerriAI/litellm, models.dev, and OpenRouter and refreshed weekly. Each row shows a per-model "verified" date. If a price is wrong, click the row to open the model page and use the inline "suggest edit" link — submissions go into a public review queue.
Which Novita AI model is cheapest?
Input pricing on Novita AI starts at $0.0100 per 1M tokens. Sort the table by price (or use the in-page filter at the top) to find the cheapest model that matches your capability requirements.
Can I route to Novita AI via an OpenAI-compatible API?
Yes — point your OpenAI client at Future AGI's Agent Command Center, configure a Novita AI target, and call Novita AI models with the standard /v1/chat/completions surface. The same gateway can route to other providers as fallback. Free for the first 100K requests/month.
Route any Novita AI model via Agent Command Center →
OpenAI-compatible endpoint. Caching, fallback, guardrails, observability. Free for 100K requests/month.