Model Comparator

Compare up to four models using the same criteria.

Hermes 4 - Llama-3.1 70B (Non-reasoning)
Nous Research
Mistral Small 3
Mistral

Differences only

✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.

Model Comparator
Metrics	Hermes 4 - Llama-3.1 70B (Non-reasoning) Nous Research	Mistral Small 3 Mistral
Intelligence	6.7	6.7
Math	Best: 11.3	4.3
Speed	—	58.4 t/s
TTFT	—	0.88 s
Input · $ / 1M tokens	—	$0.050
Output · $ / 1M tokens	—	$0.080
Context window	Best: 128K	32K
Reasoning	No	No
Open weights	Yes	Yes

Intelligence

Hermes 4 - Llama-3.1 70B (Non-reasoning): 6.7
Mistral Small 3: 6.7

Math

Hermes 4 - Llama-3.1 70B (Non-reasoning): Best: 11.3
Mistral Small 3: 4.3

Speed

Hermes 4 - Llama-3.1 70B (Non-reasoning): —
Mistral Small 3: 58.4 t/s

TTFT

Hermes 4 - Llama-3.1 70B (Non-reasoning): —
Mistral Small 3: 0.88 s

Input · $ / 1M tokens

Hermes 4 - Llama-3.1 70B (Non-reasoning): —
Mistral Small 3: $0.050

Output · $ / 1M tokens

Hermes 4 - Llama-3.1 70B (Non-reasoning): —
Mistral Small 3: $0.080

Context window

Hermes 4 - Llama-3.1 70B (Non-reasoning): Best: 128K
Mistral Small 3: 32K

Reasoning

Hermes 4 - Llama-3.1 70B (Non-reasoning): No
Mistral Small 3: No

Open weights

Hermes 4 - Llama-3.1 70B (Non-reasoning): Yes
Mistral Small 3: Yes

Each reasoning level remains a separate configuration.

Score and cost

Each line joins the reasoning levels of a compared model; the large point marks the configuration in the table. The top-right corner holds the best score for the lowest cost. Select another point to compare that level.

Mistral Small 36.7 · $0.058

Not plotted for lack of data on these axes: Hermes 4 - Llama-3.1 70B (Non-reasoning).