# Open Model API Prices by Host: gpt-oss, DeepSeek and More
Source: https://savrn.com/ai-index/pricing/open-models
Markdown alternate of the page above; the site index is https://savrn.com/llms.txt

---

[SAVRN Index.](https://savrn.com/ai-index)

Explore the IndexAI pricing

AI pricing pagesOpen model pricing

Tracking 796 prices · Checked 48 minutes ago

Open models price index −0.4%

Since Sep 11, 2026, same items only; −0.3% in 30 days. 20 price changes recorded. Typical output per million tokens $0.90 now, $1.01 on Sep 11, 2026.

[Full price history →](https://savrn.com/ai-index/pricing/history#open)

Same model, many hosts

## 44 models with more than one host

Hosts are ordered from the lowest output price. The bar under each host is its output price against the highest for that model; the spread is the highest over the lowest.

gpt-oss-120b 4.4xoutput spread OpenAI · 13 hosts · 131,072 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.037 | $0.17 | [Hugging Face](https://huggingface.co/openai/gpt-oss-120b) Oct 25, 2026 |
| Crusoe | $0.05 | $0.20 | [Research capture](https://www.crusoe.ai/cloud/pricing) Oct 21, 2026 |
| Novita | $0.05 | $0.25 | [Hugging Face](https://huggingface.co/openai/gpt-oss-120b) Oct 25, 2026 |
| Nscale | $0.10 | $0.40 | [Hugging Face](https://huggingface.co/openai/gpt-oss-120b) Oct 25, 2026 |
| OVHcloud | $0.09 | $0.47 | [Hugging Face](https://huggingface.co/openai/gpt-oss-120b) Oct 25, 2026 |
| Baseten | $0.10 | $0.50 | [Hugging Face](https://huggingface.co/openai/gpt-oss-120b) Oct 25, 2026 |
| SambaNova | $0.22 | $0.59 | [Research capture](https://cloud.sambanova.ai/plans/pricing) Oct 21, 2026 |
| Fireworks | $0.15 | $0.60 | [Provider page](https://docs.fireworks.ai/serverless/pricing) Oct 24, 2026 |
| Together | $0.15 | $0.60 | [Research capture](https://www.together.ai/pricing) Oct 21, 2026 |
| Nebius Token Factory | $0.15 | $0.60 | [Research capture](https://tokenfactory.nebius.com/endpoints) Oct 21, 2026 |
| Scaleway | $0.17 | $0.68 | [Hugging Face](https://huggingface.co/openai/gpt-oss-120b) Oct 25, 2026 |
| Groq | $0.15 | $0.75 | [Hugging Face](https://huggingface.co/openai/gpt-oss-120b) Oct 25, 2026 |
| Cerebras | $0.35 | $0.75 | [Hugging Face](https://huggingface.co/openai/gpt-oss-120b) Oct 25, 2026 |

DeepSeek-V4-Pro-0813 1.5xoutput spread DeepSeek · 5 hosts · 1,048,576 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $1.30 | $2.60 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813) Oct 25, 2026 |
| Together | $1.32 | $3.96 | [Research capture](https://www.together.ai/pricing) Oct 21, 2026 |
| Nebius Token Factory | $1.32 | $3.96 | [Research capture](https://tokenfactory.nebius.com/endpoints) Oct 21, 2026 |
| Baseten | $1.32 | $3.96 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813) Oct 25, 2026 |
| Novita | $1.32 | $3.96 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-0813) Oct 25, 2026 |

DeepSeek-V4.1-Flash 2.0xoutput spread DeepSeek · 4 hosts · 1,048,576 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.20 | $0.60 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash) Oct 25, 2026 |
| Fireworks | $0.30 | $1.20 | [Provider page](https://docs.fireworks.ai/serverless/pricing) Oct 24, 2026 |
| Baseten | $0.30 | $1.20 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash) Oct 25, 2026 |
| Novita | $0.30 | $1.20 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash) Oct 25, 2026 |

GLM-5.2 2.6xoutput spread Z.ai · 4 hosts · 1,048,576 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.75 | $2.40 | [Hugging Face](https://huggingface.co/zai-org/GLM-5.2) Oct 25, 2026 |
| Baseten | $1.40 | $4.40 | [Hugging Face](https://huggingface.co/zai-org/GLM-5.2) Oct 25, 2026 |
| Novita | $1.40 | $4.40 | [Hugging Face](https://huggingface.co/zai-org/GLM-5.2) Oct 25, 2026 |
| Scaleway | $2.05 | $6.27 | [Hugging Face](https://huggingface.co/zai-org/GLM-5.2) Oct 25, 2026 |

GLM-5.3 1.1xoutput spread Z.ai · 4 hosts · 1,048,576 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.90 | $4.00 | [Hugging Face](https://huggingface.co/zai-org/GLM-5.3) Oct 25, 2026 |
| Fireworks | $1.40 | $4.40 | [Provider page](https://docs.fireworks.ai/serverless/pricing) Oct 24, 2026 |
| Baseten | $1.40 | $4.40 | [Hugging Face](https://huggingface.co/zai-org/GLM-5.3) Oct 25, 2026 |
| Novita | $1.40 | $4.40 | [Hugging Face](https://huggingface.co/zai-org/GLM-5.3) Oct 25, 2026 |

GLM-5.3-Flash Sameprice everywhere Z.ai · 4 hosts · 1,048,576 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| Fireworks | $0.15 | $0.50 | [Provider page](https://docs.fireworks.ai/serverless/pricing) Oct 24, 2026 |
| Baseten | $0.15 | $0.50 | [Hugging Face](https://huggingface.co/zai-org/GLM-5.3-Flash) Oct 25, 2026 |
| DeepInfra | $0.15 | $0.50 | [Hugging Face](https://huggingface.co/zai-org/GLM-5.3-Flash) Oct 25, 2026 |
| Novita | $0.15 | $0.50 | [Hugging Face](https://huggingface.co/zai-org/GLM-5.3-Flash) Oct 25, 2026 |

gpt-oss-20b 3.6xoutput spread OpenAI · 4 hosts · 131,072 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.03 | $0.14 | [Hugging Face](https://huggingface.co/openai/gpt-oss-20b) Oct 25, 2026 |
| OVHcloud | $0.05 | $0.18 | [Hugging Face](https://huggingface.co/openai/gpt-oss-20b) Oct 25, 2026 |
| Nscale | $0.05 | $0.20 | [Hugging Face](https://huggingface.co/openai/gpt-oss-20b) Oct 25, 2026 |
| Groq | $0.10 | $0.50 | [Hugging Face](https://huggingface.co/openai/gpt-oss-20b) Oct 25, 2026 |

Kimi-K3 1.1xoutput spread Moonshot AI · 4 hosts · 1,048,576 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $2.85 | $14.25 | [Hugging Face](https://huggingface.co/moonshotai/Kimi-K3) Oct 25, 2026 |
| Fireworks | $3.00 | $15.00 | [Provider page](https://docs.fireworks.ai/serverless/pricing) Oct 24, 2026 |
| Together | $3.00 | $15.00 | [Research capture](https://www.together.ai/pricing) Oct 21, 2026 |
| Baseten | $3.00 | $15.00 | [Hugging Face](https://huggingface.co/moonshotai/Kimi-K3) Oct 25, 2026 |

Llama-3.3-70B-Instruct 2.6xoutput spread Meta · 4 hosts · 131,072 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| Nebius Token Factory | $0.13 | $0.40 | [Research capture](https://tokenfactory.nebius.com/endpoints) Oct 21, 2026 |
| Novita | $0.14 | $0.40 | [Hugging Face](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) Oct 25, 2026 |
| OVHcloud | $0.74 | $0.74 | [Hugging Face](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) Oct 25, 2026 |
| Scaleway | $1.03 | $1.03 | [Hugging Face](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) Oct 25, 2026 |

Qwen3.5-397B-A17B 1.4xoutput spread Qwen · 4 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.45 | $3.00 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.5-397B-A17B) Oct 25, 2026 |
| Novita | $0.60 | $3.60 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.5-397B-A17B) Oct 25, 2026 |
| Scaleway | $0.68 | $4.10 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.5-397B-A17B) Oct 25, 2026 |
| OVHcloud | $0.71 | $4.25 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.5-397B-A17B) Oct 25, 2026 |

Qwen3.8-27B 2.1xoutput spread Qwen · 4 hosts · 1,000,000 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| Cerebras | $0.99 | $1.49 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.8-27B) Oct 25, 2026 |
| DeepInfra | $0.20 | $2.50 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.8-27B) Oct 25, 2026 |
| Novita | $0.42 | $3.00 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.8-27B) Oct 25, 2026 |
| OVHcloud | $0.47 | $3.19 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.8-27B) Oct 25, 2026 |

DeepSeek-V4-Flash-0731 7.3xoutput spread DeepSeek · 3 hosts · 1,048,576 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.06 | $0.18 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731) Oct 25, 2026 |
| Baseten | $0.13 | $0.26 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731) Oct 25, 2026 |
| Novita | $0.44 | $1.32 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731) Oct 25, 2026 |

DeepSeek-V4-Pro 1.3xoutput spread DeepSeek · 3 hosts · 1,048,576 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $1.30 | $2.60 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro) Oct 25, 2026 |
| Novita | $1.60 | $3.20 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro) Oct 25, 2026 |
| Crusoe | $1.74 | $3.48 | [Research capture](https://www.crusoe.ai/cloud/pricing) Oct 21, 2026 |

DeepSeek-V4.1-Flash-1017 Sameprice everywhere DeepSeek · 3 hosts · 1,048,576 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| Fireworks | $0.30 | $1.20 | [Provider page](https://docs.fireworks.ai/serverless/pricing) Oct 24, 2026 |
| Baseten | $0.30 | $1.20 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017) Oct 25, 2026 |
| Novita | $0.30 | $1.20 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017) Oct 25, 2026 |

gemma-4-26B-A4B-it 1.7xoutput spread Google · 3 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.07 | $0.34 | [Hugging Face](https://huggingface.co/google/gemma-4-26B-A4B-it) Oct 25, 2026 |
| Novita | $0.13 | $0.40 | [Hugging Face](https://huggingface.co/google/gemma-4-26B-A4B-it) Oct 25, 2026 |
| Scaleway | $0.29 | $0.57 | [Hugging Face](https://huggingface.co/google/gemma-4-26B-A4B-it) Oct 25, 2026 |

Llama-3.1-8B-Instruct 1.2xoutput spread Meta · 3 hosts · 131,072 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.02 | $0.05 | [Hugging Face](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) Oct 25, 2026 |
| Novita | $0.02 | $0.05 | [Hugging Face](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) Oct 25, 2026 |
| Nscale | $0.06 | $0.06 | [Hugging Face](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) Oct 25, 2026 |

Llama-4-Scout-17B-16E-Instruct 2.0xoutput spread Meta · 3 hosts · 890,000 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| Nscale | $0.09 | $0.29 | [Hugging Face](https://huggingface.co/meta-llama/Llama-4-Scout-17B-16E-Instruct) Oct 25, 2026 |
| DeepInfra | $0.10 | $0.30 | [Hugging Face](https://huggingface.co/meta-llama/Llama-4-Scout-17B-16E-Instruct) Oct 25, 2026 |
| Novita | $0.18 | $0.59 | [Hugging Face](https://huggingface.co/meta-llama/Llama-4-Scout-17B-16E-Instruct) Oct 25, 2026 |

Qwen3-235B-A22B-Instruct-2507 4.7xoutput spread Qwen · 3 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.09 | $0.55 | [Hugging Face](https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507) Oct 25, 2026 |
| Nscale | $0.20 | $0.60 | [Hugging Face](https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507) Oct 25, 2026 |
| Scaleway | $0.86 | $2.57 | [Hugging Face](https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507) Oct 25, 2026 |

DeepSeek-R1-0528 1.2xoutput spread DeepSeek · 2 hosts · 163,840 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.50 | $2.15 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-R1-0528) Oct 25, 2026 |
| Novita | $0.70 | $2.50 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-R1-0528) Oct 25, 2026 |

DeepSeek-V4-Flash 5.6xoutput spread DeepSeek · 2 hosts · 1,048,576 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.09 | $0.18 | [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash) Oct 25, 2026 |
| Vultr | $0.30 | $1.00 | [Research capture](https://www.vultr.com/pricing/#cloud-inference) Oct 21, 2026 |

gemma-4-31B-it Sameprice everywhere Google · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| Novita | $0.14 | $0.40 | [Hugging Face](https://huggingface.co/google/gemma-4-31B-it) Oct 25, 2026 |
| DeepInfra | $0.20 | $0.40 | [Hugging Face](https://huggingface.co/google/gemma-4-31B-it) Oct 25, 2026 |

GLM-4.6 1.1xoutput spread Z.ai · 2 hosts · 204,800 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.50 | $2.00 | [Hugging Face](https://huggingface.co/zai-org/GLM-4.6) Oct 25, 2026 |
| Novita | $0.55 | $2.20 | [Hugging Face](https://huggingface.co/zai-org/GLM-4.6) Oct 25, 2026 |

GLM-4.7 1.3xoutput spread Z.ai · 2 hosts · 204,800 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.40 | $1.75 | [Hugging Face](https://huggingface.co/zai-org/GLM-4.7) Oct 25, 2026 |
| Novita | $0.60 | $2.20 | [Hugging Face](https://huggingface.co/zai-org/GLM-4.7) Oct 25, 2026 |

GLM-4.7-Flash Sameprice everywhere Z.ai · 2 hosts · 202,752 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.06 | $0.40 | [Hugging Face](https://huggingface.co/zai-org/GLM-4.7-Flash) Oct 25, 2026 |
| Novita | $0.07 | $0.40 | [Hugging Face](https://huggingface.co/zai-org/GLM-4.7-Flash) Oct 25, 2026 |

GLM-5 1.5xoutput spread Z.ai · 2 hosts · 202,800 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.60 | $2.08 | [Hugging Face](https://huggingface.co/zai-org/GLM-5) Oct 25, 2026 |
| Novita | $1.00 | $3.20 | [Hugging Face](https://huggingface.co/zai-org/GLM-5) Oct 25, 2026 |

Hy4-preview Sameprice everywhere Tencent · 2 hosts · 1,048,576 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.83 | $2.50 | [Hugging Face](https://huggingface.co/tencent/Hy4-preview) Oct 25, 2026 |
| Novita | $0.83 | $2.50 | [Hugging Face](https://huggingface.co/tencent/Hy4-preview) Oct 25, 2026 |

Kimi-K2.5 1.3xoutput spread Moonshot AI · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.45 | $2.25 | [Hugging Face](https://huggingface.co/moonshotai/Kimi-K2.5) Oct 25, 2026 |
| Novita | $0.60 | $3.00 | [Hugging Face](https://huggingface.co/moonshotai/Kimi-K2.5) Oct 25, 2026 |

Kimi-K2.6 Sameprice everywhere Moonshot AI · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| Novita | $0.80 | $3.40 | [Hugging Face](https://huggingface.co/moonshotai/Kimi-K2.6) Oct 25, 2026 |
| DeepInfra | $0.75 | $3.50 | [Hugging Face](https://huggingface.co/moonshotai/Kimi-K2.6) Oct 25, 2026 |

Kimi-K2.7-Code 1.2xoutput spread Moonshot AI · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.68 | $3.40 | [Hugging Face](https://huggingface.co/moonshotai/Kimi-K2.7-Code) Oct 25, 2026 |
| Novita | $0.95 | $4.00 | [Hugging Face](https://huggingface.co/moonshotai/Kimi-K2.7-Code) Oct 25, 2026 |

Ling-3.0-flash Sameprice everywhere inclusionAI · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.06 | $0.18 | [Hugging Face](https://huggingface.co/inclusionAI/Ling-3.0-flash) Oct 25, 2026 |
| Novita | $0.06 | $0.18 | [Hugging Face](https://huggingface.co/inclusionAI/Ling-3.0-flash) Oct 25, 2026 |

Ling-3.0-flash-Fin 1.2xoutput spread inclusionAI · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.06 | $0.18 | [Hugging Face](https://huggingface.co/inclusionAI/Ling-3.0-flash-Fin) Oct 25, 2026 |
| Novita | $0.075 | $0.22 | [Hugging Face](https://huggingface.co/inclusionAI/Ling-3.0-flash-Fin) Oct 25, 2026 |

Ling-3.0-flash-VL 1.2xoutput spread inclusionAI · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.06 | $0.18 | [Hugging Face](https://huggingface.co/inclusionAI/Ling-3.0-flash-VL) Oct 25, 2026 |
| Novita | $0.075 | $0.22 | [Hugging Face](https://huggingface.co/inclusionAI/Ling-3.0-flash-VL) Oct 25, 2026 |

MiniMax-M2.7 1.2xoutput spread MiniMax · 2 hosts · 204,800 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.25 | $1.00 | [Hugging Face](https://huggingface.co/MiniMaxAI/MiniMax-M2.7) Oct 25, 2026 |
| Novita | $0.30 | $1.20 | [Hugging Face](https://huggingface.co/MiniMaxAI/MiniMax-M2.7) Oct 25, 2026 |

MiniMax-M3 Sameprice everywhere MiniMax · 2 hosts · 1,000,000 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| Fireworks | $0.30 | $1.20 | [Provider page](https://docs.fireworks.ai/serverless/pricing) Oct 24, 2026 |
| Novita | $0.30 | $1.20 | [Hugging Face](https://huggingface.co/MiniMaxAI/MiniMax-M3) Oct 25, 2026 |

Qwen3-14B 1.2xoutput spread Qwen · 2 hosts · 40,960 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| Nscale | $0.07 | $0.20 | [Hugging Face](https://huggingface.co/Qwen/Qwen3-14B) Oct 25, 2026 |
| DeepInfra | $0.12 | $0.24 | [Hugging Face](https://huggingface.co/Qwen/Qwen3-14B) Oct 25, 2026 |

Qwen3-32B 1.1xoutput spread Qwen · 2 hosts · 40,960 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| Nscale | $0.08 | $0.25 | [Hugging Face](https://huggingface.co/Qwen/Qwen3-32B) Oct 25, 2026 |
| DeepInfra | $0.08 | $0.28 | [Hugging Face](https://huggingface.co/Qwen/Qwen3-32B) Oct 25, 2026 |

Qwen3-VL-30B-A3B-Instruct 1.2xoutput spread Qwen · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.15 | $0.60 | [Hugging Face](https://huggingface.co/Qwen/Qwen3-VL-30B-A3B-Instruct) Oct 25, 2026 |
| Novita | $0.20 | $0.70 | [Hugging Face](https://huggingface.co/Qwen/Qwen3-VL-30B-A3B-Instruct) Oct 25, 2026 |

Qwen3.5-122B-A10B 1.3xoutput spread Qwen · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.29 | $2.40 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.5-122B-A10B) Oct 25, 2026 |
| Novita | $0.40 | $3.20 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.5-122B-A10B) Oct 25, 2026 |

Qwen3.5-27B 1.1xoutput spread Qwen · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| Novita | $0.30 | $2.40 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.5-27B) Oct 25, 2026 |
| DeepInfra | $0.26 | $2.60 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.5-27B) Oct 25, 2026 |

Qwen3.5-35B-A3B 2.0xoutput spread Qwen · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.14 | $1.00 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.5-35B-A3B) Oct 25, 2026 |
| Novita | $0.25 | $2.00 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.5-35B-A3B) Oct 25, 2026 |

Qwen3.5-9B 1.2xoutput spread Qwen · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.10 | $0.15 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.5-9B) Oct 25, 2026 |
| OVHcloud | $0.12 | $0.18 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.5-9B) Oct 25, 2026 |

Qwen3.6-27B Sameprice everywhere Qwen · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| OVHcloud | $0.47 | $3.19 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.6-27B) Oct 25, 2026 |
| DeepInfra | $0.32 | $3.20 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.6-27B) Oct 25, 2026 |

Qwen3.6-35B-A3B 1.8xoutput spread Qwen · 2 hosts · 262,144 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $0.10 | $0.95 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.6-35B-A3B) Oct 25, 2026 |
| Scaleway | $0.29 | $1.71 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.6-35B-A3B) Oct 25, 2026 |

Qwen3.8-2.4T-A95B Sameprice everywhere Qwen · 2 hosts · 1,000,000 tokens context

| Host | Input | Output | Source |
| --- | --- | --- | --- |
| DeepInfra | $2.00 | $6.00 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B) Oct 25, 2026 |
| Novita | $2.00 | $6.00 | [Hugging Face](https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B) Oct 25, 2026 |

One host each

## 75 models priced by one host

| Model | Host | Input | Output | Other published prices | Checked |
| --- | --- | --- | --- | --- | --- |
| ALIA-40b-instruct-2606BSC-LT | Public AI | $0.30 | $0.60 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/BSC-LT/ALIA-40b-instruct-2606) |
| Apertus-70B-Instruct-2509Swiss AI | Public AI | $0.82 | $2.92 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/swiss-ai/Apertus-70B-Instruct-2509) |
| Apertus-8B-Instruct-2509Swiss AI | Public AI | $0.10 | $0.20 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/swiss-ai/Apertus-8B-Instruct-2509) |
| Apertus-v1.5-70BSwiss AI | Public AI | $0.82 | $2.92 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/swiss-ai/Apertus-v1.5-70B) |
| AutoGLM-Phone-9B-MultilingualZ.ai | Novita | $0.035 | $0.14 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/zai-org/AutoGLM-Phone-9B-Multilingual) |
| Beam-1.0-501B-A23Breflection | Baseten | $0.40 | $1.20 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/reflection/Beam-1.0-501B-A23B) |
| Bielik-11B-v3.0-InstructSpeakLeash | Public AI | $0.40 | $0.40 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/speakleash/Bielik-11B-v3.0-Instruct) |
| DeepSeek-R1DeepSeek | Novita | $0.70 | $2.50 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/deepseek-ai/DeepSeek-R1) |
| DeepSeek-R1-Distill-Llama-8BDeepSeek | Nscale | $0.05 | $0.05 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) |
| DeepSeek-R1-Distill-Qwen-14BDeepSeek | Nscale | $0.20 | $0.20 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) |
| DeepSeek-R1-Distill-Qwen-7BDeepSeek | Nscale | $0.15 | $0.15 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B) |
| DeepSeek-V3DeepSeek | DeepInfra | $0.32 | $0.89 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/deepseek-ai/DeepSeek-V3) |
| DeepSeek-V3-0324DeepSeek | DeepInfra | $0.24 | $0.90 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/deepseek-ai/DeepSeek-V3-0324) |
| DeepSeek-V3.1DeepSeek | DeepInfra | $0.25 | $0.95 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/deepseek-ai/DeepSeek-V3.1) |
| DeepSeek-V3.1-TerminusDeepSeek | DeepInfra | $0.27 | $0.95 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus) |
| DeepSeek-V3.2DeepSeek | DeepInfra | $0.26 | $0.38 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/deepseek-ai/DeepSeek-V3.2) |
| DeepSeek-V4-Flash-Vision-ExpDeepSeek | DeepInfra | $0.44 | $1.32 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-Vision-Exp) |
| Ember-1 | Fireworks | $3.00 | $15.00 | Priority $3.75 / $18.75 | On pageOct 24, 2026[docs.fireworks.ai ↗](https://docs.fireworks.ai/serverless/pricing) |
| Flash | DeepSeek | $0.30 | $1.20 | None published | ResearchOct 21, 2026[api-docs.deepseek.com ↗](https://api-docs.deepseek.com/quick_start/pricing/) |
| gemma-3-12b-itGoogle | DeepInfra | $0.05 | $0.15 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/google/gemma-3-12b-it) |
| gemma-3-27b-itGoogle | DeepInfra | $0.08 | $0.16 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/google/gemma-3-27b-it) |
| gemma-3-4b-itGoogle | DeepInfra | $0.05 | $0.10 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/google/gemma-3-4b-it) |
| Gemma-SEA-LION-v4-27B-ITAI Singapore | Public AI | $0.20 | $0.40 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/aisingapore/Gemma-SEA-LION-v4-27B-IT) |
| GLM-4.5-AirZ.ai | Novita | $0.13 | $0.85 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/zai-org/GLM-4.5-Air) |
| GLM-4.5VZ.ai | Novita | $0.60 | $1.80 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/zai-org/GLM-4.5V) |
| GLM-4.6V-FlashZ.ai | Novita | $0.30 | $0.90 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/zai-org/GLM-4.6V-Flash) |
| GLM-5.1Z.ai | DeepInfra | $1.05 | $3.50 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/zai-org/GLM-5.1) |
| granite-4.2-30bIBM | DeepInfra | $0.16 | $0.65 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/ibm-granite/granite-4.2-30b) |
| granite-4.2-3bIBM | DeepInfra | $0.03 | $0.12 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/ibm-granite/granite-4.2-3b) |
| granite-4.2-8bIBM | DeepInfra | $0.06 | $0.25 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/ibm-granite/granite-4.2-8b) |
| Hermes-3-Llama-3.1-70BNous Research | DeepInfra | $0.70 | $0.70 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-70B) |
| Hy3Tencent | DeepInfra | $0.13 | $0.53 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/tencent/Hy3) |
| InklingThinking Machines | DeepInfra | $0.95 | $4.05 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/thinkingmachines/Inkling) |
| Inkling-SmallThinking Machines | DeepInfra | $0.45 | $1.20 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/thinkingmachines/Inkling-Small) |
| Kimi-K2-InstructMoonshot AI | Novita | $0.57 | $2.30 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/moonshotai/Kimi-K2-Instruct) |
| Kimi-K2-Instruct-0905Moonshot AI | Novita | $0.60 | $2.50 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905) |
| L3-8B-Stheno-v3.2Sao10K | Novita | $0.05 | $0.05 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Sao10K/L3-8B-Stheno-v3.2) |
| Large 3 | Mistral | $0.50 | $1.50 | None published | ResearchOct 21, 2026[mistral.ai ↗](https://mistral.ai/pricing/api/) |
| Ling-3.1-flashinclusionAI | Novita | $0.30 | $0.90 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/inclusionAI/Ling-3.1-flash) |
| Llama-4-Maverick-17B-128E-Instruct-FP8Meta | Novita | $0.27 | $0.85 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8) |
| MiMo-V2.5Xiaomi MiMo | DeepInfra | $0.14 | $0.28 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/XiaomiMiMo/MiMo-V2.5) |
| MiMo-V2.5-ProXiaomi MiMo | DeepInfra | $1.00 | $3.00 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/XiaomiMiMo/MiMo-V2.5-Pro) |
| MiMo-V2.6-Flash-RLXiaomi MiMo | Novita | $0.14 | $0.28 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL) |
| MiMo-V2.6-Pro-RLXiaomi MiMo | Novita | $0.44 | $0.87 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL) |
| MiniMax-M1-80kMiniMax | Novita | $0.55 | $2.20 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/MiniMaxAI/MiniMax-M1-80k) |
| MiniMax-M2.5MiniMax | Novita | $0.30 | $1.20 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/MiniMaxAI/MiniMax-M2.5) |
| Muse Spark Openmeta-models | Fireworks | $1.25 | $4.25 | Priority $1.88 / $6.38 | On pageOct 24, 2026[docs.fireworks.ai ↗](https://docs.fireworks.ai/serverless/pricing) |
| Muse-Glimmer-30Bmeta-models | DeepInfra | $0.30 | $1.20 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/meta-models/Muse-Glimmer-30B) |
| Muse-Spark-Open-816B-A42Bmeta-models | Baseten | $1.25 | $4.25 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/meta-models/Muse-Spark-Open-816B-A42B) |
| Nemotron-SEA-LION-v4.8-120B-A12BAI Singapore | Public AI | $0.50 | $2.00 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/aisingapore/Nemotron-SEA-LION-v4.8-120B-A12B) |
| NVIDIA Nemotron 3 Ultra (Preview)NVIDIA | Fireworks | $0.60 | $2.40 | Priority $0.75 / $3.00 | On pageOct 24, 2026[docs.fireworks.ai ↗](https://docs.fireworks.ai/serverless/pricing) |
| NVIDIA Nemotron 3.5 Lightning 30B A3BNVIDIA | Fireworks | $0.05 | $0.20 | Priority $0.0625 / $0.25 | On pageOct 24, 2026[docs.fireworks.ai ↗](https://docs.fireworks.ai/serverless/pricing) |
| NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16NVIDIA | DeepInfra | $1.00 | $5.00 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16) |
| NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4NVIDIA | Baseten | $0.20 | $0.50 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4) |
| phi-4Microsoft | DeepInfra | $0.07 | $0.14 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/microsoft/phi-4) |
| Qwen 3.8 MaxQwen | Fireworks | $2.00 | $6.00 | Priority $3.00 / $9.00 | On pageOct 24, 2026[docs.fireworks.ai ↗](https://docs.fireworks.ai/serverless/pricing) |
| Qwen-SEA-LION-v4-32B-ITAI Singapore | Public AI | $0.25 | $0.50 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/aisingapore/Qwen-SEA-LION-v4-32B-IT) |
| Qwen-SEA-LION-v4.5-27B-ITAI Singapore | Public AI | $0.20 | $0.40 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/aisingapore/Qwen-SEA-LION-v4.5-27B-IT) |
| Qwen2.5-72B-InstructQwen | DeepInfra | $0.36 | $0.40 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Qwen/Qwen2.5-72B-Instruct) |
| Qwen2.5-Coder-32B-InstructQwen | Nscale | $0.06 | $0.20 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Qwen/Qwen2.5-Coder-32B-Instruct) |
| Qwen2.5-Coder-3B-InstructQwen | Nscale | $0.01 | $0.03 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Qwen/Qwen2.5-Coder-3B-Instruct) |
| Qwen2.5-Coder-7B-InstructQwen | Nscale | $0.01 | $0.03 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Qwen/Qwen2.5-Coder-7B-Instruct) |
| Qwen2.5-VL-72B-InstructQwen | OVHcloud | $1.01 | $1.01 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Qwen/Qwen2.5-VL-72B-Instruct) |
| Qwen3-235B-A22BQwen | Nscale | $0.20 | $0.60 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Qwen/Qwen3-235B-A22B) |
| Qwen3-235B-A22B-Thinking-2507Qwen | DeepInfra | $0.23 | $2.30 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507) |
| Qwen3-30B-A3BQwen | DeepInfra | $0.12 | $0.50 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Qwen/Qwen3-30B-A3B) |
| Qwen3-4B-Instruct-2507Qwen | Nscale | $0.01 | $0.03 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507) |
| Qwen3-4B-Thinking-2507Qwen | Nscale | $0.01 | $0.03 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Qwen/Qwen3-4B-Thinking-2507) |
| Qwen3-8BQwen | Nscale | $0.07 | $0.18 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Qwen/Qwen3-8B) |
| Qwen3-Next-80B-A3B-InstructQwen | DeepInfra | $0.09 | $1.10 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Instruct) |
| Qwen3-VL-235B-A22B-InstructQwen | DeepInfra | $0.20 | $0.88 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/Qwen/Qwen3-VL-235B-A22B-Instruct) |
| Step-3.5-FlashStepFun | DeepInfra | $0.10 | $0.30 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/stepfun-ai/Step-3.5-Flash) |
| Step-3.7-FlashStepFun | DeepInfra | $0.20 | $1.15 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/stepfun-ai/Step-3.7-Flash) |
| V4 Pro | DeepSeek | $1.32 | $3.96 | None published | ResearchOct 21, 2026[api-docs.deepseek.com ↗](https://api-docs.deepseek.com/quick_start/pricing/) |
| WizardLM-2-8x22Balpindale | Novita | $0.62 | $0.62 | None published | FeedOct 25, 2026[huggingface.co ↗](https://huggingface.co/alpindale/WizardLM-2-8x22B) |

The SAVRN reference set

## Seven references, one system

Capital, delay, policy, water, grid, permits and scarcity are one system seen from seven sides. Each reference is primary-sourced, refreshed on a schedule, and links into the others, so a project can be followed from the money committed to the docket that stopped it.

Capital Atlas

Who is funding AI data centers: capital providers, named sites, financing transactions, and the blocked-and-delayed record, state by state and company by company.

[Open the atlas →](https://savrn.com/data-center-capital-atlas)

Delay Watchlist

Project-level slippage: named projects, jurisdictional actions, utility and federal dockets, SEC disclosures, and the interconnection queue.

Open the watchlist →

Moratorium Tracker

The policy layer: every state measure and local moratorium mapped and cited, with one page per state.

Open the tracker →

Water Tracker

Every federal, state, local and interstate water measure that touches a data center, plus a fifty-state water ledger.

Open the water tracker →

Grid Operator Watchlist

ERCOT, PJM, MISO, CAISO, SPP, NYISO, ISO-NE and FERC: the large-load rules, filings and queue positions that decide whether a site can energize.

Open the watchlist →

Permits and Power Development

Data center and substation permits, large loads in the interconnection queues, and the power plants planned to serve them, refreshed hourly from public records.

Open the permits tracker →

SAVRN Scarcity Tracker

Eleven primary-sourced indicators of AI infrastructure scarcity, one composite index, read against the token index, SAVRN's own cost basis and the frontier ceiling.

Open the scarcity tracker →

Have a specific address? Site Intelligence reads all seven references against one location, county, utility and grid operator included, and cites every finding.

[Check an address →](https://savrn.com/site-intelligence)