Compare/Devstral Small (May '25) vs Qwen3.5 2B (Reasoning)

Devstral Small (May '25)vsQwen3.5 2B (Reasoning)

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

Mistral

Devstral Small (May '25)

Input
$0.06/M
Output
$0.12/M
Speed
TTFT
Alibaba

Qwen3.5 2B (Reasoning)

Input
$0.02/M
Output
$0.1/M
Speed
357 tok/s
TTFT
0.54s

Winner by Category

Cheaper
Qwen3.5 2B (Reasoning)
Faster (tok/s)
Qwen3.5 2B (Reasoning)
Lower Latency
Devstral Small (May '25)
Benchmarks (9-1)
Devstral Small (May '25)

Pricing Comparison

MetricDevstral Small (May '25)Qwen3.5 2B (Reasoning)
Input ($/M tokens)$0.06$0.02
Output ($/M tokens)$0.12$0.1
Cost for 1M input + 100K output tokens:
Devstral Small (May '25)$0.07
Qwen3.5 2B (Reasoning)$0.03

Speed Comparison

Output Speed (tokens/s) — higher is better
Devstral Small (May '25)
Qwen3.5 2B (Reasoning)
357 tok/s
Time to First Token (seconds) — lower is better
Devstral Small (May '25)
Qwen3.5 2B (Reasoning)
0.54s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
18.016.3
Coding Index
12.23.5
Math Index
GPQA Diamond
43.4%45.6%
MMLU-Pro
63.2%
LiveCodeBench
25.8%
AIME 2025
MATH-500
68.4%
Humanity's Last Exam
4.0%2.1%
SciCode
24.5%2.8%
IFBench
31.6%31.5%
TerminalBench
6.1%3.8%
Devstral Small (May '25)9 wins
1 winsQwen3.5 2B (Reasoning)

Frequently Asked Questions

Which is cheaper, Devstral Small (May '25) or Qwen3.5 2B (Reasoning)?

Qwen3.5 2B (Reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.04/M tokens vs $0.07/M for Devstral Small (May '25).

Which model performs better on benchmarks?

Devstral Small (May '25) wins 9 out of 12 benchmarks compared to 1 for Qwen3.5 2B (Reasoning). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

Qwen3.5 2B (Reasoning) generates tokens faster at 357 tok/s vs 0 tok/s. Devstral Small (May '25) also has lower time-to-first-token (0.00s vs 0.54s).

When should I use Devstral Small (May '25) vs Qwen3.5 2B (Reasoning)?

Choose based on your priorities: Qwen3.5 2B (Reasoning) for lower cost, Devstral Small (May '25) for stronger benchmark performance, and Qwen3.5 2B (Reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.