tinyanalysis

Compare / Ministral 3 8B vs DeepSeek-R1-0528-Qwen3-8B

Ministral 3 8B vs DeepSeek-R1-0528-Qwen3-8B

Updated November 2026

Verdict

Ministral 3 8B scores higher on the Small LM Index: 55.2 vs 53.1, although their likely ranges overlap, so the lead is not certain. On the 3 benchmarks both labs report, Ministral 3 8B wins 3 and DeepSeek-R1-0528-Qwen3-8B wins 0. Both need about 6.8 GB of memory at 4-bit quantization. Only Ministral 3 8B has a hosted API on OpenRouter ($0.15 per million tokens); the other has to be self-hosted.

Pick Ministral 3 8B if…

  • you want the strongest model overall
  • you mostly write code
  • you need math and reasoning
  • you want a hosted API rather than running it yourself
  • you work with long documents
  • you need image input

Pick DeepSeek-R1-0528-Qwen3-8B if…

  • you want the more widely used model (more quantized builds and community support)

Specifications

Ministral 3 8BDeepSeek-R1-0528-Qwen3-8B
Small LM Index55.2 (likely 51–60)53.1 (likely 48–58)
LabMistral AI · EuropeDeepSeek · China
Released2025-12-022025-05-29
Parameters8B (dense)8.2B (dense)
Memory at 4-bit~6.7 GB~6.8 GB
Context window256K tokens128K tokens
Inputtext, imagetext
ReasoningHybrid (thinking on/off)Always thinks
LicenseApache-2.0MIT
API price / 1M tokens$0.15 ($0.15 in, $0.15 out)Self-host only
Hugging Face downloads (30 days)16.3K357.7K

Benchmarks

Higher is better. Bold marks the winner where both labs report a score. ≈ marks the index model's estimate for a score that wasn't reported, ✓ an independent measurement and * a score measured by another lab.

BenchmarkMinistral 3 8BDeepSeek-R1-0528-Qwen3-8B
GPQA Diamond Knowledge66.861.1
AIME Math78.776.3
LiveCodeBench Coding61.660.5

Scores are mostly reported by each lab with its own settings, so small gaps are noise. See the methodology. Open in the interactive comparison to add more models.

Questions

Which is better, Ministral 3 8B or DeepSeek-R1-0528-Qwen3-8B?

Ministral 3 8B scores higher on the Small LM Index: 55.2 vs 53.1, although their likely ranges overlap, so the lead is not certain. On the 3 benchmarks both labs report, Ministral 3 8B wins 3 and DeepSeek-R1-0528-Qwen3-8B wins 0.

Which is cheaper to run?

Only Ministral 3 8B has a hosted API on OpenRouter ($0.15 per million tokens); the other has to be self-hosted. Both need about 6.8 GB of memory at 4-bit quantization.

Which has the longer context window?

Ministral 3 8B has the longer context window: 256K tokens against 128K.

Can I use Ministral 3 8B and DeepSeek-R1-0528-Qwen3-8B commercially?

Ministral 3 8B is released under Apache-2.0 and DeepSeek-R1-0528-Qwen3-8B under MIT. Apache 2.0 and MIT allow commercial use; other licenses can add conditions, so read the license before shipping a product.

Related comparisons

Guides featuring these models