GLM-4.5 vs Hermes 3 405B Instruct

Prices checked

GLM-4.5 takes 2 of 5 rounds against Hermes 3 405B Instruct. It is also the cheaper of the two to run, at $0.35 per million tokens against $1.

Z.aiWins 2 rounds

GLM-4.5

131k context · tool calling · open weights

$0.35/M tokens

Cheapest price, at Submodel · $0.17 per 1,000 pages

Nous Research

Hermes 3 405B Instruct

131k context · open weights

$1/M tokens

OpenRouter's own price · $0.66 per 1,000 pages

Maker's price in (lower wins)

$0.60/M

$1/M

Maker's price out (lower wins)

$2.20/M

$1/M

Cheapest price (lower wins)

$0.35/M at Submodel

none below the maker

Maker's price, blended (lower wins)

$1/M

$1/M

Cost per 1,000 pages (lower wins)

$0.17

$0.66

Context

131k tokens

131k tokens

Open weights

yes

yes

Tool calling

yes

no

Structured output

not published

yes

Reads images

no

no

Reasoning

yes

no

Released

28 Jul 2025

16 Aug 2024

Providers selling it

23

9

Price we measure against

the maker's own

OpenRouter

Trains on your prompts (no is better)

not published

no provider to check

Providers selling both · 4

  • DeepInfra$0.70 / $1
  • Eden AI$0.70 / $1
  • Kilo Gateway$1 / $1
  • OpenRouter$1 / $1

GLM-4.5, then Hermes 3 405B Instruct, per million tokens.

Go further

Add a third or fourth model, or change the weighting, in the tool.

Frequently asked

Which is cheaper, GLM-4.5 or Hermes 3 405B Instruct?
GLM-4.5, at $0.35 per million tokens against $1 for Hermes 3 405B Instruct, blended 3 in : 1 out, read on 31 Oct 2026.
Where is each one cheapest?
GLM-4.5: $0.35 at Submodel. Hermes 3 405B Instruct: nobody sells it below OpenRouter's own price with a stated reason, so $1 is the number to beat.
Can I buy both from one provider?
Yes. 4 providers sell both: DeepInfra, Eden AI, Kilo Gateway, OpenRouter. The cheapest of them for the two together is DeepInfra.
Which holds more context?
Both hold 131k tokens in one call.

Prices exclude tax, read 31 Oct 2026 from each provider's own catalog. They are per M tokens, blended 3 in : 1 out, and a page is 600 tokens in and 60 out. Bars are scaled to the larger value in each row.

Related comparisons

More