> ## Documentation Index
> Fetch the complete documentation index at: https://docs.sideren.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Max plan

> What the Max plan costs, how much usage it gives, its 5-hour and weekly limits and the 127 models it includes.

* **Price:** \$49 a month
* **Usage:** Up to \$250 of usage a month\*
* **5-hour limit:** 30% of the month's usage
* **Weekly limit:** 60% of the month's usage
* **Models:** 127, [listed below](#models-included)

\* Based on DeepSeek V4.1 Flash 1017, with cached input counted at its normal input price. Real usage depends on the model, context and caching.

The month's usage renews with each billing cycle and doesn't carry over. At 100% of a limit, requests pause until it resets; with extra usage on, they carry on using purchased credit. See [Usage limits](https://docs.sideren.io/plans/usage-limits.md).

Subscribe or upgrade on the [pricing page](https://www.sideren.io/pricing). An upgrade applies at once, for the prorated difference; a downgrade at your next renewal. See [Payment methods](https://docs.sideren.io/plans/payment-methods.md).

## Models included

Max adds 26 models to Pro's: 25 from OpenAI and 1 from Anthropic, among them Claude Opus 5.5 (`opus-5.5`) and GPT-6 Astra (`gpt-6-astra`).

Prices are list prices in US dollars per 1M tokens. A dash under Cached input means the model has no cached-input discount: cached input costs the input price.

### Anthropic

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Claude Opus 5.5 `opus-5.5` | 1M | \$4 | \$0.2 | \$20 |
| Claude Sonnet 5.5 `sonnet-5.5` | 1M | \$2 | \$0.2 | \$10 |
| Claude Haiku 5.5 `haiku-5.5` | 100K | \$0.1 | \$0.01 | \$0.5 |

### OpenAI

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| GPT-6 Astra `gpt-6-astra` | 272K | \$10 | \$1 | \$50 |
| GPT-6.1 Sol `gpt-6.1-sol` | 272K | \$2 | \$0.1 | \$10 |
| GPT-6 Sol `gpt-6-sol` | 272K | \$2 | \$0.2 | \$10 |
| GPT-5.6 Sol `gpt-5.6-sol` | 272K | \$5 | \$0.5 | \$30 |
| GPT-5.5 `gpt-5.5` | 272K | \$5 | \$0.5 | \$30 |
| GPT-6.1 Terra `gpt-6.1-terra` | 272K | \$1 | \$0.05 | \$5 |
| GPT-5 Codex `gpt-5-codex` | 400K | \$1.25 | \$0.125 | \$10 |
| GPT-5.6 Terra `gpt-5.6-terra` | 272K | \$2 | \$0.2 | \$12 |
| GPT-5.4 `gpt-5.4` | 272K | \$2.5 | \$0.25 | \$15 |
| GPT-5.3 Codex `gpt-5.3-codex` | 400K | \$1.75 | \$0.175 | \$14 |
| GPT-5.2 `gpt-5.2` | 400K | \$1.75 | \$0.175 | \$14 |
| GPT-5.2 Codex `gpt-5.2-codex` | 400K | \$1.75 | \$0.175 | \$14 |
| GPT-5.1 `gpt-5.1` | 400K | \$1.25 | \$0.125 | \$10 |
| GPT-5.1 Codex Max `gpt-5.1-codex-max` | 400K | \$1.25 | \$0.125 | \$10 |
| GPT-5.1 Codex `gpt-5.1-codex` | 400K | \$1.25 | \$0.125 | \$10 |
| GPT-5 `gpt-5` | 400K | \$1.25 | \$0.125 | \$10 |
| GPT-5.4 mini `gpt-5.4-mini` | 400K | \$0.75 | \$0.075 | \$4.5 |
| GPT-6 Luna `gpt-6-luna` | 272K | \$0.1 | \$0.01 | \$0.5 |
| GPT-5.6 Luna `gpt-5.6-luna` | 272K | \$0.2 | \$0.02 | \$1.2 |
| GPT-5.1 Codex mini `gpt-5.1-codex-mini` | 400K | \$0.25 | \$0.025 | \$2 |
| GPT Chat (latest) `gpt-chat-latest` | 200K | \$5 | \$0.5 | \$30 |
| GPT-5 mini `gpt-5-mini` | 400K | \$0.25 | \$0.025 | \$2 |
| GPT-5.4 nano `gpt-5.4-nano` | 400K | \$0.2 | \$0.02 | \$1.25 |
| GPT-4.1 `gpt-4.1` | 1.05M | \$2 | \$0.5 | \$8 |
| GPT-5 nano `gpt-5-nano` | 400K | \$0.05 | \$0.005 | \$0.4 |
| GPT-OSS 120B `@sdn/gpt-oss-120b` | 128K | \$0.35 | — | \$0.75 |
| GPT-4.1 mini `gpt-4.1-mini` | 1.05M | \$0.4 | \$0.1 | \$1.6 |
| GPT-OSS Safeguard 120B `@sdn/gpt-oss-safeguard-120b` | 128K | \$0.15 | — | \$0.6 |
| GPT-4.1 nano `gpt-4.1-nano` | 1.05M | \$0.1 | \$0.025 | \$0.4 |
| GPT-OSS Safeguard 20B `@sdn/gpt-oss-safeguard-20b` | 128K | \$0.07 | — | \$0.2 |
| GPT-OSS 20B `@sdn/gpt-oss-20b` | 128K | \$0.2 | — | \$0.3 |

`gpt-chat-latest` retires on Dec 2, 2026.

### Moonshot AI

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Kimi K3.1 Pro `@sdn/kimi-k3.1-pro` | 1.05M | \$2 | \$0.2 | \$10 |
| Kimi K3 `@sdn/kimi-k3` | 1M | \$3 | \$0.3 | \$15 |
| Kimi K2.7 Code `@sdn/kimi-k2.7-code` | 131K | \$0.95 | \$0.19 | \$4 |
| Kimi K2.6 `@sdn/kimi-k2.6` | 262K | \$0.95 | \$0.16 | \$4 |
| Kimi K2.5 `@sdn/kimi-k2.5` | 262K | \$0.6 | — | \$3 |
| Kimi K2 Thinking `@sdn/kimi-k2-thinking` | 262K | \$0.6 | — | \$2.5 |

### Z.ai

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| GLM 5.3 `@sdn/glm-5.3` | 1M | \$1.4 | \$0.26 | \$4.4 |
| GLM 5.2 `@sdn/glm-5.2` | 131K | \$1.4 | \$0.26 | \$4.4 |
| GLM 5.3 Flash `@sdn/glm-5.3-flash` | 1M | \$0.15 | \$0.03 | \$0.5 |
| GLM 5.3 Flash Ultra Fast `@sdn/glm-5.3-flash-ultra-fast` | 1M | \$0.3 | \$0.06 | \$1 |
| GLM 5 `@sdn/glm-5` | 200K | \$1 | — | \$3.2 |
| GLM 4.7 `@sdn/glm-4.7` | 200K | \$0.6 | — | \$2.2 |
| GLM 4.6 `@sdn/glm-4.6` | 200K | \$0.6 | — | \$2.2 |
| GLM 4.7 Flash `@sdn/glm-4.7-flash` | 131K | \$0.0605 | — | \$0.4 |

`@sdn/glm-5.3-flash-ultra-fast` is 50% off at launch, from Nov 8, 2026, 17:00 UTC until Nov 15, 2026, 17:00 UTC. The table shows its list price, which applies after.

### DeepSeek

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| DeepSeek V4 Pro `@sdn/deepseek-v4-pro` | 1M | \$1.32 | \$0.044 | \$3.96 |
| DeepSeek V4 Flash `@sdn/deepseek-v4-flash` | 1M | \$0.44 | \$0.014 | \$1.32 |
| DeepSeek V4.1 Flash 1017 `@sdn/deepseek-v4.1-flash-1017` | 1M | \$0.3 | \$0.006 | \$1.2 |
| DeepSeek V4.1 Flash Ultra Fast `@sdn/deepseek-v4.1-flash-ultra-fast` | 1M | \$0.6 | \$0.012 | \$2.4 |
| DeepSeek V3.2 `@sdn/deepseek-v3.2` | 131K | \$0.57 | \$0.114 | \$1.71 |
| DeepSeek V3.1 `@sdn/deepseek-v3.1` | 164K | \$0.58 | — | \$1.68 |
| DeepSeek R1 `@sdn/deepseek-r1` | 128K | \$1.35 | — | \$5.4 |

`@sdn/deepseek-v4.1-flash-ultra-fast` is 50% off at launch, from Nov 8, 2026, 17:00 UTC until Nov 15, 2026, 17:00 UTC. The table shows its list price, which applies after.

### xAI

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Grok 4.7 `grok-4.7` | 200K | \$2 | \$0.5 | \$6 |
| Grok 4.6 `grok-4.6` | 500K | \$2 | \$0.5 | \$6 |
| Grok 4.3 `grok-4.3` | 1M | \$1.25 | \$0.2 | \$2.5 |
| Grok 4 `grok-4` | 256K | \$3 | — | \$15 |
| Grok 4.20 `grok-4.20` | 2M | \$1.25 | — | \$2.5 |
| Grok 4.20 (non-reasoning) `grok-4.20-non-reasoning` | 2M | \$1.25 | — | \$2.5 |
| Grok 4.1 Fast `grok-4.1-fast` | 2M | \$0.2 | — | \$0.5 |

### Meta

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Muse Spark 1.3 `muse-spark-1.3` | 1.05M | \$1.25 | \$0.15 | \$4.25 |
| Muse Spark 1.3 Contributor `muse-spark-1.3-contributor` | 1.05M | \$0.1 | \$0.002 | \$0.2 |
| Llama 4 Maverick `@sdn/llama-4-maverick` | 1M | \$0.24 | — | \$0.97 |
| Llama 3.3 70B `@sdn/llama-3.3-70b` | 128K | \$0.72 | — | \$0.72 |
| Llama 3.1 70B `@sdn/llama-3.1-70b` | 128K | \$0.72 | — | \$0.72 |
| Llama 3.2 11B Vision `@sdn/llama-3.2-11b-vision` | 128K | \$0.049 | — | \$0.676 |
| Llama 4 Scout `@sdn/llama-4-scout` | 3.5M | \$0.17 | — | \$0.66 |
| Llama 3.2 3B `@sdn/llama-3.2-3b` | 80K | \$0.051 | — | \$0.335 |
| Llama 3 70B `@sdn/llama-3-70b` | 8K | \$2.65 | — | \$3.5 |
| Llama 3.1 8B `@sdn/llama-3.1-8b` | 128K | \$0.22 | — | \$0.22 |
| Llama 3.2 1B `@sdn/llama-3.2-1b` | 60K | \$0.027 | — | \$0.201 |
| Llama 3 8B `@sdn/llama-3-8b` | 8K | \$0.3 | — | \$0.6 |

`muse-spark-1.3-contributor`: Meta may use prompts and replies to train its models. A key with zero data retention can't use it.

### Alibaba Qwen

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Qwen3 Coder 480B `@sdn/qwen3-coder-480b` | 131K | \$0.45 | \$0.045 | \$1.8 |
| Qwen3 235B `@sdn/qwen3-235b` | 262K | \$0.22 | \$0.022 | \$0.88 |
| Qwen 3.8 27B `@sdn/qwen3.8-27b` | 262K | \$0.45 | — | \$3.2 |
| Qwen3 Coder Next `@sdn/qwen3-coder-next` | 262K | \$0.5 | — | \$1.2 |
| Qwen3 VL 235B `@sdn/qwen3-vl-235b` | 262K | \$0.3 | — | \$1.5 |
| Qwen3 Next 80B `@sdn/qwen3-next-80b` | 262K | \$0.14 | — | \$1.2 |
| Qwen3 Coder 30B `@sdn/qwen3-coder-30b` | 262K | \$0.15 | — | \$0.6 |
| Qwen3 30B A3B `@sdn/qwen3-30b-a3b` | 33K | \$0.051 | — | \$0.335 |
| Qwen3 32B `@sdn/qwen3-32b` | 16K | \$0.15 | — | \$0.6 |
| Qwen2.5 Coder 32B `@sdn/qwen2.5-coder-32b` | 33K | \$0.66 | — | \$1 |

### Cohere

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Command A `@sdn/command-a` | 256K | \$2.5 | — | \$10 |

### MiniMax

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| MiniMax M3.1 Flash `@sdn/minimax-m3.1-flash` | 300K | \$0.15 | \$0.015 | \$0.6 |
| MiniMax M2.5 `@sdn/minimax-m2.5` | 197K | \$0.3 | — | \$1.2 |
| MiniMax M2.1 `@sdn/minimax-m2.1` | 205K | \$0.3 | — | \$1.2 |
| MiniMax M2 `@sdn/minimax-m2` | 205K | \$0.3 | — | \$1.2 |

### Sarvam

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Sarvam 105B `@sdn/sarvam-105b` | 128K | \$0.18 | \$0.06 | \$0.7 |
| Sarvam 105B Conversations `sarvam-105b-conversations` | 32K | \$0.18 | \$0.06 | \$0.7 |

### Mistral AI

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Mistral Medium 3.5 `mistral-medium-3.5` | 256K | \$1.5 | — | \$7.5 |
| Devstral 2 123B `@sdn/devstral-2-123b` | 262K | \$0.4 | — | \$2 |
| Mistral Large 3 675B `@sdn/mistral-large-3-675b` | 262K | \$0.5 | — | \$1.5 |
| Magistral Small `@sdn/magistral-small` | 128K | \$0.5 | — | \$1.5 |
| Mistral Small 3.1 `@sdn/mistral-small-3.1` | 131K | \$0.351 | — | \$0.555 |
| Pixtral Large `@sdn/pixtral-large` | 128K | \$2 | — | \$6 |
| Mistral Large 24.02 `@sdn/mistral-large-2402` | 33K | \$4 | — | \$12 |
| Ministral 14B `@sdn/ministral-14b` | 128K | \$0.2 | — | \$0.2 |
| Mistral Small `@sdn/mistral-small` | 33K | \$1 | — | \$3 |
| Ministral 8B `@sdn/ministral-8b` | 128K | \$0.15 | — | \$0.15 |
| Ministral 3B `@sdn/ministral-3b` | 256K | \$0.1 | — | \$0.1 |

### NVIDIA

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Nemotron 3 120B `@sdn/nemotron-3-120b` | 256K | \$0.5 | — | \$1.5 |
| Nemotron Super 3 120B `@sdn/nemotron-super-3-120b` | 262K | \$0.15 | — | \$0.65 |
| Nemotron Nano 3 30B `@sdn/nemotron-nano-3-30b` | 262K | \$0.06 | — | \$0.24 |
| Nemotron Nano 12B `@sdn/nemotron-nano-12b` | 128K | \$0.2 | — | \$0.6 |

### Microsoft

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Phi-4 `@sdn/phi-4` | 16K | \$0.125 | — | \$0.5 |
| Phi-4 mini `@sdn/phi-4-mini` | 131K | \$0.075 | — | \$0.3 |

### Google

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Gemma 4 26B `@sdn/gemma-4-26b` | 256K | \$0.1 | — | \$0.3 |
| Gemma 3 27B `@sdn/gemma-3-27b` | 203K | \$0.12 | — | \$0.2 |
| Gemma 3 12B `@sdn/gemma-3-12b` | 131K | \$0.05 | — | \$0.1 |
| Gemma 3 4B `@sdn/gemma-3-4b` | 128K | \$0.04 | — | \$0.08 |

### Amazon

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Nova Pro `nova-pro` | 300K | \$0.8 | \$0.2 | \$3.2 |
| Nova 2 Lite `nova-2-lite` | 128K | \$0.33 | — | \$2.75 |
| Nova Lite `nova-lite` | 300K | \$0.06 | \$0.015 | \$0.24 |
| Nova Micro `nova-micro` | 128K | \$0.035 | \$0.0088 | \$0.14 |

### Writer

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Palmyra X5 `palmyra-x5` | 1M | \$0.6 | — | \$6 |
| Palmyra X4 `palmyra-x4` | 123K | \$2.5 | — | \$10 |

### AI Singapore

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Gemma Sea Lion v4 27B `@sdn/gemma-sea-lion-v4-27b` | 128K | \$0.351 | — | \$0.555 |

### IBM

| Model | Context | Input | Cached input | Output |
| - | - | -: | -: | -: |
| Granite 4.0 H Micro `@sdn/granite-4.0-h-micro` | 131K | \$0.017 | — | \$0.112 |

### Embedding models

Text in, vectors out ([Embeddings](https://docs.sideren.io/api/embeddings.md)). They have no output price.

| Model | Max input | Dimensions | Input |
| - | - | -: | -: |
| Qwen3 Embedding 0.6B `@sdn/qwen3-embed-0.6b` | 8K | 1,024 | \$0.02 |
| BGE-M3 `@sdn/bge-m3` | 8K | 1,024 | \$0.02 |
| BGE Large EN v1.5 `@sdn/bge-large-en` | 1K | 1,024 | \$0.21 |
| EmbeddingGemma 300M `@sdn/embed-gemma-300m` | 2K | 768 | \$0.02 |
| BGE Base EN v1.5 `@sdn/bge-base-en` | 1K | 768 | \$0.07 |
| PLaMo Embedding 1B `@sdn/plamo-embed-1b` | 4K | 2,048 | \$0.02 |
| BGE Small EN v1.5 `@sdn/bge-small-en` | 1K | 384 | \$0.03 |


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.