Compare/Mistral Small 3 vs Olmo 3 7B Instruct

Mistral Small 3vsOlmo 3 7B Instruct

Side-by-side comparison of pricing, 12 benchmarks, and generation speed.

Mistral

Mistral Small 3

Input
$0.075/M
Output
$0.19/M
Speed
154 tok/s
TTFT
0.52s
Allen Institute for AI

Olmo 3 7B Instruct

Input
$0.1/M
Output
$0.2/M
Speed
TTFT

Winner by Category

Cheaper
Mistral Small 3
Faster (tok/s)
Mistral Small 3
Lower Latency
Olmo 3 7B Instruct
Benchmarks (5-7)
Olmo 3 7B Instruct

Pricing Comparison

MetricMistral Small 3Olmo 3 7B Instruct
Input ($/M tokens)$0.075$0.1
Output ($/M tokens)$0.19$0.2
Cost for 1M input + 100K output tokens:
Mistral Small 3$0.09
Olmo 3 7B Instruct$0.12

Speed Comparison

Output Speed (tokens/s) — higher is better
Mistral Small 3
154 tok/s
Olmo 3 7B Instruct
Time to First Token (seconds) — lower is better
Mistral Small 3
0.52s
Olmo 3 7B Instruct

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
12.78.1
Coding Index
3.4
Math Index
4.341.3
GPQA Diamond
46.2%40.0%
MMLU-Pro
65.2%52.2%
LiveCodeBench
25.2%26.6%
AIME 2025
4.3%41.3%
MATH-500
71.5%
Humanity's Last Exam
4.1%5.8%
SciCode
23.6%10.3%
IFBench
26.4%32.8%
TerminalBench
0.0%
Mistral Small 35 wins
7 winsOlmo 3 7B Instruct

Frequently Asked Questions

Which is cheaper, Mistral Small 3 or Olmo 3 7B Instruct?

Mistral Small 3 is cheaper overall. Its blended price (3:1 input/output ratio) is $0.10/M tokens vs $0.13/M for Olmo 3 7B Instruct.

Which model performs better on benchmarks?

Olmo 3 7B Instruct wins 7 out of 12 benchmarks compared to 5 for Mistral Small 3. See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

Mistral Small 3 generates tokens faster at 154 tok/s vs 0 tok/s. However, Olmo 3 7B Instruct has lower time-to-first-token (0.00s vs 0.52s).

When should I use Mistral Small 3 vs Olmo 3 7B Instruct?

Choose based on your priorities: Mistral Small 3 for lower cost, Olmo 3 7B Instruct for stronger benchmark performance, and Mistral Small 3 for faster generation. For latency-sensitive apps, check the TTFT comparison above.