DeepSeek-R1 0528 vs Magistral Medium 1
Wins 4 of 6 areas
Coding · Reasoning · Long documents · Following instructions
Wins 0 of 6 areas
—
DeepSeek-R1 0528 is the stronger all-rounder.
Scores updated · 20 tests both models report · How we compare
Where each one wins
Tests won in each of the six areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- ReasoningHard problems that need careful thinking30DeepSeek-R1 05283 of 3 tests
- CodingWriting and fixing software20DeepSeek-R1 05282 of 2 tests
- Long documentsFinding answers in very long texts10DeepSeek-R1 05281 of 1 test
- Following instructionsDoing exactly what it is asked10DeepSeek-R1 05281 of 1 test
- AgentsCarrying out multi-step tasks on its own00Even0 each · 1 tie
- FactsGetting facts right instead of making them up11Even1 each
Agents, long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where DeepSeek-R1 0528 pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+14.5points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+13.4points ahead
- Code for real scientific research problemsSciCode+10.6points ahead
Where Magistral Medium 1 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+24.1points ahead
Every test, side by side
All 20 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDeepSeek-R1 0528
- SciCodeDeepSeek-R1 0528 by 10.640.329.7+10.6
- Terminal-Bench HardDeepSeek-R1 0528 by 6.815.99.1+6.8
ReasoningDeepSeek-R1 0528
- GPQA DiamondDeepSeek-R1 0528 by 13.481.367.9+13.4
- Humanity's Last ExamDeepSeek-R1 0528 by 615.89.8+6
- CritPtDeepSeek-R1 0528 by 1.11.40.3+1.1
FactsEven
- AA-Omniscience · Non-hallucinationMagistral Medium 1 by 24.116.640.7+24.1
- AA-Omniscience · AccuracyDeepSeek-R1 0528 by 10.230.520.3+10.2
Following instructionsDeepSeek-R1 0528
- IFBenchDeepSeek-R1 0528 by 14.539.625.1+14.5
Other results10 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- Aider-PolyglotDeepSeek-R1 0528 by 24.571.647.1+24.5
- LiveCodeBenchDeepSeek-R1 0528 by 2373.350.3+23
- AIME 2025DeepSeek-R1 0528 by 22.687.564.9+22.6
- AIME 2024DeepSeek-R1 0528 by 20.791.470.7+20.7
- ARC-AGI-1DeepSeek-R1 0528 by 15.121.26.1+15.1
- τ²-Bench Telecom (AA run)DeepSeek-R1 0528 by 13.436.523.1+13.4
- Artificial Analysis Coding IndexDeepSeek-R1 0528 by 82416+8
- AA Agentic IndexDeepSeek-R1 0528 by 5.120.815.7+5.1
- AA IntelligenceDeepSeek-R1 0528 by 413.19.1+4
- AA-Omnisciencetie-27.4-26.9tie
Questions people ask
Which is better, DeepSeek-R1 0528 or Magistral Medium 1?
DeepSeek-R1 0528 wins four of the six areas where both have results: coding, reasoning, long documents and following instructions. Magistral Medium 1 wins none. They are level on agents and facts.
Which is better for coding?
DeepSeek-R1 0528. It wins 2 of the 2 coding tests both models report; Magistral Medium 1 wins none.
How do you compare the two?
We use the 20 benchmark tests both models have published scores on. The verdict counts the 10 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 10 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.