---
title: "Hermes 4 405B API pricing"
canonical: https://uplinkswitch.ai/models/hermes-4-405b
generated: 2026-10-20T19:38:52.465Z
source: https://api.uplinkswitch.com/v1/catalog
base_url: https://api.uplinkswitch.com/v1
---

# Hermes 4 405B API pricing

Hermes 4 405B API at $0.90/M input and $3.60/M output tokens. Same SDK, one key, one URL.

## Price per million tokens (USD)

| Tokens | Nous Research list price | UplinkSwitch | Under list |
| --- | --- | --- | --- |
| Input | — | $0.90 | — |
| Cached input | — | billed at the input rate | — |
| Output | — | $3.60 | — |

## Model

- Model id: `hermes-4-405b`
- Formats: OpenAI SDK
- Base URL: https://api.uplinkswitch.com/v1
- Context window: 128,000 tokens
- Input: text
- Daily real-model check: Same fingerprint since Oct 20 (https://uplinkswitch.ai/proof#hermes-4-405b)

## Request

### Python · OpenAI SDK

```python
from openai import OpenAI

client = OpenAI(
    base_url="https://api.uplinkswitch.com/v1",
    api_key="sk-uplink-live-...",
)

completion = client.chat.completions.create(
    model="hermes-4-405b",
    messages=[{"role": "user", "content": "Summarize this in one line."}],
)
print(completion.choices[0].message.content)
```

### TypeScript · OpenAI SDK

```typescript
import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://api.uplinkswitch.com/v1',
  apiKey: 'sk-uplink-live-...',
});

const completion = await client.chat.completions.create({
  model: 'hermes-4-405b',
  messages: [{ role: 'user', content: 'Summarize this in one line.' }],
});
console.log(completion.choices[0].message.content);
```

### curl

```bash
curl https://api.uplinkswitch.com/v1/chat/completions \
  -H "authorization: Bearer sk-uplink-live-..." \
  -H "content-type: application/json" \
  -d '{"model":"hermes-4-405b","messages":[{"role":"user","content":"Hello"}]}'
```

## Questions

### What is Hermes 4 405B?

Hermes 4 405B is a model from Nous Research, with a 128,000-token context window, taking text input. Through UplinkSwitch it answers the OpenAI SDK on one key as hermes-4-405b.

### How much does the Hermes 4 405B API cost?

$0.90 per million input tokens and $3.60 per million output tokens.

### Is Hermes 4 405B free?

It is pay as you go: prepaid credit from $10, with no subscription or monthly fee, billed per token. A 1,000-token prompt with a 500-token reply costs about $0.0027.

### What is the Hermes 4 405B context window?

Hermes 4 405B has a 128,000-token context window. It takes text input.

### What is the Hermes 4 405B model id?

hermes-4-405b. Send it as the model through the OpenAI SDK with base URL https://api.uplinkswitch.com/v1.

### How do I get a Hermes 4 405B API key?

Create one in the console at uplinkswitch.ai/console: prepaid credit from $10, no subscription, and the key is live at once. The same key calls Hermes 4 405B and every other model in the catalog; send it as the bearer token with base URL https://api.uplinkswitch.com/v1.

### Is it the real Hermes 4 405B?

Every day each model gets the same test prompt, and a model always counts that prompt the same way; a swapped one would not. Hermes 4 405B's latest result: Same fingerprint since Oct 20.

### Is there a Hermes 4 405B API?

Yes: hermes-4-405b on the OpenAI format at https://api.uplinkswitch.com/v1, $0.90 input and $3.60 output per million tokens. No weights to download and no GPU of your own; the key also runs every other model in the catalog.

### What is Hermes 4 405B based on?

Meta’s Llama 3.1 405B. Nous Research post-trained it on about 5 million samples (about 60 billion tokens) of reasoning and non-reasoning data, per the Hermes 4 405B card.

### What license does Hermes 4 405B have?

The card’s license tag is llama3, Meta’s Llama community license, which Hermes inherits from Llama 3.1 405B.

### What does Hermes 4 405B cost per 100 messages?

$0.36 for 100 chat turns of 2,000 input and 500 output tokens each, at $0.90 input and $3.60 output per million tokens, from prepaid credit with no subscription.

### Is Hermes 4 405B uncensored?

Nous Research trains Hermes to follow the user and the system prompt rather than a fixed policy; its card describes it as aligned to you, with broadly neutral alignment. It is a post-train, not an abliteration. For builds with the refusal direction removed from the weights, the uncensored list at uplinkswitch.ai/uncensored marks them Abliterated.

### Does Hermes 4 405B reason?

Yes, when it decides to: its card calls it a hybrid-mode reasoning model that writes its deliberation in <think> segments, and answers directly otherwise. Give it room in max_tokens (1,000 or more) for problems that need working out. At $0.90 input and $3.60 output per million tokens, thinking tokens bill as output.

### Hermes 4 405B or Hermes 3 Llama 3.1 70B?

Hermes 4 is the 405B reasoning model; Hermes 3 is the 70B generalist from the previous series. Hermes 4 405B is $0.90 / $3.60 per million tokens in / out, $3.60 for 1,000 chat turns; Hermes 3 Llama 3.1 70B is $1.224 / $1.224 per million tokens in / out, $3.06 for 1,000 chat turns. Both run on the same key, so send the same prompt to each.

## Related

- Uncensored LLM API: https://uplinkswitch.ai/uncensored
- SillyTavern setup: https://uplinkswitch.ai/guides/sillytavern
- Janitor AI setup: https://uplinkswitch.ai/guides/janitor-ai
- Chub AI setup: https://uplinkswitch.ai/guides/chub-ai
- SDKs setup: https://uplinkswitch.ai/guides/sdks
- Every model and price: https://uplinkswitch.ai/llms-full.txt
- Machine catalog: https://uplinkswitch.ai/ai-catalog.json