Compare/Mistral Medium 3.5 vs Qwen3.6 Max Preview

Mistral Medium 3.5vsQwen3.6 Max Preview

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

Mistral

Mistral Medium 3.5

Input
$1.5/M
Output
$7.5/M
Speed
163 tok/s
TTFT
0.53s
Alibaba

Qwen3.6 Max Preview

Input
$1.3/M
Output
$7.8/M
Speed
33 tok/s
TTFT
2.15s

Winner by Category

Cheaper
Qwen3.6 Max Preview
Faster (tok/s)
Mistral Medium 3.5
Lower Latency
Mistral Medium 3.5
Benchmarks (0-7)
Qwen3.6 Max Preview

Pricing Comparison

MetricMistral Medium 3.5Qwen3.6 Max Preview
Input ($/M tokens)$1.5$1.3
Output ($/M tokens)$7.5$7.8
Cost for 1M input + 100K output tokens:
Mistral Medium 3.5$2.25
Qwen3.6 Max Preview$2.08

Speed Comparison

Output Speed (tokens/s) — higher is better
Mistral Medium 3.5
163 tok/s
Qwen3.6 Max Preview
33 tok/s
Time to First Token (seconds) — lower is better
Mistral Medium 3.5
0.53s
Qwen3.6 Max Preview
2.15s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
39.251.8
Coding Index
35.444.9
Math Index
GPQA Diamond
74.8%88.8%
MMLU-Pro
LiveCodeBench
AIME 2025
MATH-500
Humanity's Last Exam
12.8%28.9%
SciCode
39.6%46.9%
IFBench
68.8%76.6%
TerminalBench
33.3%43.9%
Mistral Medium 3.50 wins
7 winsQwen3.6 Max Preview

Frequently Asked Questions

Which is cheaper, Mistral Medium 3.5 or Qwen3.6 Max Preview?

Qwen3.6 Max Preview is cheaper overall. Its blended price (3:1 input/output ratio) is $2.92/M tokens vs $3.00/M for Mistral Medium 3.5.

Which model performs better on benchmarks?

Qwen3.6 Max Preview wins 7 out of 12 benchmarks compared to 0 for Mistral Medium 3.5. See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

Mistral Medium 3.5 generates tokens faster at 163 tok/s vs 33 tok/s. Mistral Medium 3.5 also has lower time-to-first-token (0.53s vs 2.15s).

When should I use Mistral Medium 3.5 vs Qwen3.6 Max Preview?

Choose based on your priorities: Qwen3.6 Max Preview for lower cost, Qwen3.6 Max Preview for stronger benchmark performance, and Mistral Medium 3.5 for faster generation. For latency-sensitive apps, check the TTFT comparison above.