Mistral Devstral Small 1.1
No content available for this output.
Mistral Devstral Small 1.1 vs OpenAI o3
Compare Mistral Devstral Small 1.1 by Mistral AI against OpenAI o3 by OpenAI, tested across 47 shared challenges. Updated February 2026.
FAQ
Loading share button...
Mistral Devstral Small 1.1vsOpenAI o3
42 challenges
Why Mistral Devstral Small 1.1?
127.3x cheaper overall$0.10/M in · $0.30/M out
3 months newerJul 2025
Why OpenAI o3?
Leads 1 of 1 benchmarks
Stronger on SWE-bench Verified69.1% vs 46.8%
Favorites
Mistral Devstral Small 1.1OpenAI o3
12 / 42
Estimate Complexityreasoning
Tests an AI's ability to make educated estimates based on technical knowledge
OpenAI o3
Below is a simple "back-of-the-envelope" estimate that uses only a few public facts about GPT-3 and very generic rules of thumb for counting operations.
Step 1. Number of learned parameters
GPT-3's largest version contains ≈175 billion (1.75 × 10¹¹) trainable parameters.
Step 2. How much data was seen during training
OpenAI reports that the model was trained on about 300 billion tokens (3 × 10¹¹).
Step 3. How many floating-point operations does one parameter cost per token?
A practical rule of thumb for Transformer language models is
Training FLOPs ≈ 6 × (# parameters) × (# training tokens)
Where does the factor 6 come from?
• Forward pass: every parameter is involved in one multiply-accumulate → 2 FLOPs
• Back-prop gradient computation: roughly another 2 FLOPs
• Parameter-update (optimizer) and miscellaneous overhead: ≈2 FLOPs
Total ≈6 FLOPs per parameter per token.
Step 4. Plug in the numbers
Total FLOPs = 6 × (1.75 × 10¹¹) × (3 × 10¹¹)
= 6 × 5.25 × 10²²
≈ 3.15 × 10²³ FLOPs.
Step 5. Express the answer in more familiar units
3.15 × 10²³ ≈ 315 zettaFLOPs (1 ZFLOP = 10²¹ FLOPs).
Step 6. Sanity check with hardware-time intuition
An NVIDIA A100 delivers ≈ 312 teraFLOP/s (0.312 PFLOP/s) in mixed precision.
Wall-clock time on a single A100:
3.15 × 10²³ / 3.12 × 10¹⁴ ≈ 1.0 × 10⁹ s ≈ 32 years.
Running on 1,000 A100s in parallel would take ≈12 days, which is the right ballpark for a large industrial training run.
Final estimate
Training GPT-3 (175 B parameters, 300 B tokens) requires on the order of 3 × 10²³ floating-point operations, i.e. roughly 300 zettaFLOPs.
Vote
2 / 42
AI Board Game Logicreasoning
Tests an AI's ability to understand game rules and strategy
Mistral Devstral Small 1.1
No content available for this output.
OpenAI o3
Vote
3 / 42
Logic Puzzlereasoning
Tests an AI's ability to solve a simple but potentially confusing logic puzzle
Mistral Devstral Small 1.1
No content available for this output.
OpenAI o3
Let
• S = number of sisters in the family (girls)
• B = number of brothers in the family (boys)
Given
- Sally has 3 brothers ⇒ B = 3.
- Each brother has 2 sisters ⇒ every boy sees exactly 2 girls in the sibling group.
Therefore S = 2.
Since Sally herself is one of those 2 sisters, the number of sisters Sally has (other than herself) is
2 – 1 = 1.
Sally has 1 sister.
Vote
4 / 42
Math Misconception Testreasoning
Tests an AI's understanding of number representation
Mistral Devstral Small 1.1
No content available for this output.
OpenAI o3
No content available for this output.
Vote
5 / 42
Stochastic Consistency Testanalysis
Tests an AI's randomness and creativity
Mistral Devstral Small 1.1
No content available for this output.
OpenAI o3
No content available for this output.
Vote
6 / 42
SVG Layout Challengeimage generation
Tests an AI's ability to generate vector graphics
Mistral Devstral Small 1.1
No content available for this output.
OpenAI o3
No content available for this output.
Vote
7 / 42
Xbox Controller SVG Artimage generation
Tests an AI's ability to create detailed SVG illustrations of gaming hardware
Mistral Devstral Small 1.1
No content available for this output.
OpenAI o3
No content available for this output.
Vote
8 / 42
Realistic AI Interviewconversation
Tests an AI's ability to simulate personalities and predict future trends
Mistral Devstral Small 1.1
No content available for this output.
OpenAI o3
No content available for this output.
Vote
9 / 42
Satirical Fake News Headlineconversation
Tests an AI's humor and understanding of current events
Mistral Devstral Small 1.1
No content available for this output.
OpenAI o3
No content available for this output.
Vote
10 / 42
Character Voice Testconversation
Tests an AI's ability to write in distinct character voices
Mistral Devstral Small 1.1
No content available for this output.
OpenAI o3
No content available for this output.
Vote
11 / 42
Minimalist Landing Pageweb design
Tests an AI's ability to generate a complete, working landing page
Mistral Devstral Small 1.1
OpenAI o3
Vote
12 / 42
Pokémon Battle UI Recreationweb design
Recreate an interactive, nostalgic Pokémon battle UI in a single HTML file.
Mistral Devstral Small 1.1
No content available for this output.
OpenAI o3
No content available for this output.
Vote
12 of 42