Side-by-side comparison of pricing, 12 benchmarks, and generation speed.
| Metric | Qwen3.5 35B A3B (Non-reasoning) | Mistral Medium 3 |
|---|---|---|
| Input ($/M tokens) | $0.25 | $0.4 |
| Output ($/M tokens) | $2 | $2 |
Verdict. Qwen3.5 35B A3B (Non-reasoning) wins the overall benchmark matchup 2–0 across 2 overlapping categories, but raw benchmark score is only one input to the decision.
Pricing. Both models sit in the budget bracket for output-token pricing. At 1.0× the per-million-token cost, Mistral Medium 3 is meaningfully cheaper if your traffic is output-heavy (long completions, document generation, agent loops). Mistral Medium 3 makes more sense when output volume is low and absolute reasoning quality justifies the premium.
Strengths. Qwen3.5 35B A3B (Non-reasoning) is strongest on Coding Index (37.0), Intelligence Index (24.3). Mistral Medium 3 leads on Intelligence Index (12.5).
Speed. Throughput is comparable — 164 tok/s vs 132 tok/s — so generation speed shouldn't drive your choice here. Look at the per-benchmark wins instead.
Provider. Alibaba and Mistral sell to overlapping but distinct developer audiences: Alibaba tends to ship frontier reasoning models with premium positioning, while Mistral often prices more aggressively. Your existing vendor relationships, billing, and SLA preferences may matter as much as the raw numbers above.
Workload cost. Workload scenarios (per million requests at 30M input + 15M output tokens): Qwen3.5 35B A3B (Non-reasoning) costs $37.50 ($450/year); Mistral Medium 3 costs $42.00 ($504/year). At a smaller 5M-input/2M-output scale (single-developer tool or prototype): Qwen3.5 35B A3B (Non-reasoning) ≈ $5.25/run, Mistral Medium 3 ≈ $6.00/run. At agent/realtime scale (200M input / 100M output per million requests): Qwen3.5 35B A3B (Non-reasoning) ≈ $250/run, Mistral Medium 3 ≈ $280/run. Qwen3.5 35B A3B (Non-reasoning) becomes more attractive at higher volume — the absolute per-token pricing difference compounds when you ship at scale.
Recommendation. Both models have legitimate use cases — the right answer depends on whether you are optimizing for benchmark ceiling, latency, or unit cost. Start with the cheaper / faster model, evaluate against your specific task, and only switch if the upgrade shows a meaningful lift.
Data from Artificial Analysis API — 12 benchmarks
Qwen3.5 35B A3B (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $0.69/M tokens vs $0.80/M for Mistral Medium 3.
Qwen3.5 35B A3B (Non-reasoning) wins 2 out of 12 benchmarks compared to 0 for Mistral Medium 3. See the detailed benchmark chart above for per-category results.
Qwen3.5 35B A3B (Non-reasoning) generates tokens faster at 164 tok/s vs 132 tok/s. Qwen3.5 35B A3B (Non-reasoning) also has lower time-to-first-token (2.15s vs 2.28s).
Choose based on your priorities: Qwen3.5 35B A3B (Non-reasoning) for lower cost, Qwen3.5 35B A3B (Non-reasoning) for stronger benchmark performance, and Qwen3.5 35B A3B (Non-reasoning) for faster generation. For latency-sensitive apps, check the TTFT comparison above.