Price Per TokenPrice Per Token
Nousresearch
Nousresearch
vs
Nousresearch
Nousresearch

DeepHermes 3 Mistral 24B Preview vs Hermes 4 405B

A detailed comparison of pricing, benchmarks, and capabilities

OpenClaw

Best LLMs for OpenClaw Vote for which model works best with OpenClaw

112 out of our 301 tracked models have had a price change in February.

Get our weekly newsletter on pricing changes, new releases, and tools.

Key Takeaways

DeepHermes 3 Mistral 24B Preview wins:

  • Cheaper input tokens
  • Cheaper output tokens

Hermes 4 405B wins:

  • Larger context window
  • Faster response time
  • Higher intelligence benchmark
  • Better at coding
  • Better at math
  • Supports tool calls
Price Advantage
DeepHermes 3 Mistral 24B Preview
Benchmark Advantage
Hermes 4 405B
Context Window
Hermes 4 405B
Speed
Hermes 4 405B

Pricing Comparison

Price Comparison

MetricDeepHermes 3 Mistral 24B PreviewHermes 4 405BWinner
Input (per 1M tokens)$0.02$1.00 DeepHermes 3 Mistral 24B Preview
Output (per 1M tokens)$0.10$3.00 DeepHermes 3 Mistral 24B Preview
Cache Read (per 1M)$10000.00N/A DeepHermes 3 Mistral 24B Preview
Using a 3:1 input/output ratio, DeepHermes 3 Mistral 24B Preview is 97% cheaper overall.

DeepHermes 3 Mistral 24B Preview Providers

Chutes $0.02 (Cheapest)

Hermes 4 405B Providers

Nebius $1.00 (Cheapest)

Benchmark Comparison

7
Benchmarks Compared
0
DeepHermes 3 Mistral 24B Preview Wins
4
Hermes 4 405B Wins

Benchmark Scores

BenchmarkDeepHermes 3 Mistral 24B PreviewHermes 4 405BWinner
Intelligence Index
Overall intelligence score
10.917.6
Coding Index
Code generation & understanding
-18.1-
Math Index
Mathematical reasoning
-15.3-
MMLU Pro
Academic knowledge
58.072.9
GPQA
Graduate-level science
38.253.6
LiveCodeBench
Competitive programming
19.554.6
AIME
Competition math
4.7--
Hermes 4 405B significantly outperforms in coding benchmarks.

Cost vs Quality

X-axis:
Y-axis:
Loading chart...
DeepHermes 3 Mistral 24B Preview
Other models

Context & Performance

Context Window

DeepHermes 3 Mistral 24B Preview
32,768
tokens
Max output: 32,768 tokens
Hermes 4 405B
131,072
tokens
Hermes 4 405B has a 75% larger context window.

Speed Performance

MetricDeepHermes 3 Mistral 24B PreviewHermes 4 405BWinner
Tokens/second0.0 tok/s32.1 tok/s
Time to First Token0.00s0.75s
Hermes 4 405B responds Infinity% faster on average.

Capabilities

Feature Comparison

FeatureDeepHermes 3 Mistral 24B PreviewHermes 4 405B
Vision (Image Input)
Tool/Function Calls
Reasoning Mode
Audio Input
Audio Output
PDF Input
Prompt Caching
Web Search

License & Release

PropertyDeepHermes 3 Mistral 24B PreviewHermes 4 405B
LicenseOpen SourceOpen Source
AuthorNousresearchNousresearch
ReleasedMay 2025Aug 2025

DeepHermes 3 Mistral 24B Preview Modalities

Input
text
Output
text

Hermes 4 405B Modalities

Input
text
Output
text

Related Comparisons

Compare DeepHermes 3 Mistral 24B Preview with:

Compare Hermes 4 405B with:

Frequently Asked Questions

DeepHermes 3 Mistral 24B Preview has cheaper input pricing at $0.02/M tokens. DeepHermes 3 Mistral 24B Preview has cheaper output pricing at $0.10/M tokens.
Hermes 4 405B scores higher on coding benchmarks with a score of 18.1, compared to DeepHermes 3 Mistral 24B Preview's score of N/A.
DeepHermes 3 Mistral 24B Preview has a 32,768 token context window, while Hermes 4 405B has a 131,072 token context window.
DeepHermes 3 Mistral 24B Preview does not support vision. Hermes 4 405B does not support vision.