Decision support

DeepSeek R1 0528 Qwen3 8B vs Magistral Small

Compare canonical objects directly. The useful dimensions change with the object type, so model and hardware comparisons stay separate.

Benchmark results

Head to head on 2 benchmarks both models report

  • DeepSeek R1 0528 Qwen3 8B1lead
  • Magistral Small1lead
  • ARC-AGI-2

    Best published score · Higher is better

    • DeepSeek R1 0528 Qwen3 8B
      Not reported
    • Magistral Small
      0.0%
      #78 of 85
    Runs and sources
    DeepSeek R1 0528 Qwen3 8B

    Not reported

    Magistral Small
    • default0.0%
    Source:External evaluation via Epoch AIOriginal-source terms apply
  • GPQA diamond

    Best published score · Higher is better

    • DeepSeek R1 0528 Qwen3 8B
      9.3%
      #161 of 161
    • Magistral Small
      Best56.1%
      +46.8 pts ahead · #128 of 161
    Runs and sources
    DeepSeek R1 0528 Qwen3 8B
    • default9.3%
    Magistral Small
    • default56.1%
  • OTIS Mock AIME 2024-2025

    Best published score · Higher is better

    • DeepSeek R1 0528 Qwen3 8B
      Best43.9%
      +13.9 pts ahead · #123 of 158
    • Magistral Small
      30.0%
      #128 of 158
    Runs and sources
    DeepSeek R1 0528 Qwen3 8B
    • default43.9%
    Magistral Small
    • default30.0%

Each model's best published result is shown; "Not reported" means no matched evaluation, not a score of zero. Arena and DeepSWE ranks are the leaderboards' own; other ranks are among catalogue models with a score. Results from different harnesses or settings are not directly comparable.

Original-source terms apply to evaluations credited to their original publishers.

Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.

Model specifications

Release date
May 29, 2025
June 10, 2025
Context window
131K
131K
Native context
—
—
Max output tokens
32,000
8,192
Knowledge cutoff
—
—
Open weights
Open weights
Open weights
Parameter count
8B
23.6B

API pricing · USD per 1M tokens

Input
—
—
Output
—
—
Cached input
—
—
Cache write
—
—

Capabilities and modalities

Input modalities
text
text
Output modalities
text
text
Tool calling
—
Supported
Reasoning option
Supported
Supported
Supported options
reasoning, temperature
tools, reasoning, temperature
Capabilities
—
—

Architecture and availability

Architecture
—
—
Model families
—
—
Tokenizer
—
—
Instruction format
—
—
License
MIT
Apache 2.0
Providers
0
0
Provider endpoints
0
0
Ollama variants
0
0
Metadata source
ai-research
ai-research

A dash means the field is not reported. Provider-specific performance and pricing may differ.

To give you the best experience, we use cookies for personalization, insights, and performance. Learn more.