We use cookies. Cookies and browser storage keep you signed in and remember your preferences. With your permission we also measure how the site is used. You can any time. See our cookie policy.

New here? See how it works A 1-minute tour of the playground, models and your API key.

Glimpse vs Hermes 4 405B: API price, context & features · api14

Elorian: Glimpse vs Nous: Hermes 4 405B

A side-by-side comparison of Elorian: Glimpse ($2.10 / 1M tokens) and Nous: Hermes 4 405B ($3.50 / 1M tokens) using current USD prices and capabilities from the api14 catalog. Both run through the same OpenAI-compatible API.

Specifications

	Elorian: Glimpse	Nous: Hermes 4 405B
Provider	Elorian	Nous
Input	$2.10 / 1M tokens	$3.50 / 1M tokens
Output	$10.50 / 1M tokens	$10.50 / 1M tokens
Cache read	$0.21 / 1M tokens	$0.70 / 1M tokens
Context window	262,144	131,072
Max output	256,000	117,964
Input	text, image, video	text
Output	text	text
Tool calling
Structured output
Reasoning control
Added	Oct 17, 2026	Aug 26, 2025

Key differences

Elorian: Glimpse is 40% cheaper on input tokens.
Elorian: Glimpse reads 2.0× more context (262,144 tokens).
Only Elorian: Glimpse supports tool calling.
A request with 1,000 input and 500 output tokens costs $0.00735 with Elorian: Glimpse and $0.00875 with Nous: Hermes 4 405B.

In depth

Elorian: Glimpse is a text model from Elorian, and Nous: Hermes 4 405B is a text model from Nous. Both are listed on api14 under the text category, and both output plain text. The practical difference starts with what you can send in: Glimpse accepts text, image and video input, while Hermes 4 405B accepts text only. Glimpse also carries a much larger context window at 262144 tokens against 131072 for Hermes 4 405B, so it can hold roughly twice as much material in a single request.

On capabilities, the two overlap more than they diverge. Both expose reasoning and include_reasoning, both support structured outputs and response_format, and both accept the usual sampling controls such as temperature, top_p, top_k, frequency_penalty, presence_penalty, repetition_penalty, logit_bias, logprobs and top_logprobs. The clear split is tooling: Glimpse lists tools and tool_choice, while Hermes 4 405B does not, so function calling and agent style workflows belong on the Elorian side. Hermes 4 405B lists seed, which Glimpse does not, and that matters if you want repeatable sampling. Glimpse is the cheaper of the two on input, while output pricing is about the same for both.

For day to day work, pick based on what you are feeding the model and what it has to do with the result. If your input is text only and you want a straightforward generation or reasoning call, Hermes 4 405B is a reasonable default, and its seed parameter helps when you need the same output twice. Choose Glimpse when the request includes images or video, when the prompt plus retrieved context runs long enough that 131072 tokens is tight, or when the model needs to call tools. The larger context and the multimodal input are the two reasons that most often force the decision, and the lower input price makes that choice easier to justify at volume.

Calling either model through api14 works the same way, because the API is OpenAI compatible. You point your client at the api14 base URL, send your key, and set the model field to elorian/glimpse or nousresearch/hermes-4-405b. Everything else in the request body stays as you already write it: messages, temperature, max_tokens, response_format and the rest. Switching between the two is a one line change to the model id, so you can keep a single code path and route per request. Just remember that tools and tool_choice only apply to Glimpse, and seed only applies to Hermes 4 405B, so guard those fields if you share one request builder.

There are still cases where Hermes 4 405B is the better pick even though Glimpse is larger and cheaper on input. If your workload is pure text and fits comfortably inside 131072 tokens, the extra context of Glimpse buys you nothing, and the seed parameter gives you a reproducibility control that Glimpse does not offer. For batch jobs where you want to rerun the same prompt and compare outputs, that alone can decide it.

Switch with one string

Both models use the same endpoint and request shape. Change the model value from "elorian/glimpse" to "nousresearch/hermes-4-405b" and prompts, streaming and tools keep working.

Elorian: Glimpse details & playground →Nous: Hermes 4 405B details & playground →

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api14.com/api/v1",
  apiKey: process.env.API14_API_KEY,
});

const result = await client.chat.completions.create({
  "model": "nousresearch/hermes-4-405b",
  "messages": [
    {
      "role": "user",
      "content": "Explain vector databases in two sentences."
    }
  ]
});
console.log(result.choices[0].message.content);

Elorian: Glimpse vs Nous: Hermes 4 405B

Specifications

Key differences

In depth

Switch with one string

More comparisons