DeepSeek-V4-Flash vs North-Mini-Code-1.0
Wins 5 of 6 areas
Coding · Agents · Reasoning · Long documents · Following instructions
Wins 0 of 6 areas
—
DeepSeek-V4-Flash is the stronger all-rounder.
Scores updated · 21 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.
- CodingWriting and fixing software51DeepSeek-V4-Flash5 of 6 tests
- AgentsCarrying out multi-step tasks on its own30DeepSeek-V4-Flash3 of 3 tests
- ReasoningHard problems that need careful thinking30DeepSeek-V4-Flash3 of 3 tests
- Long documentsFinding answers in very long texts10DeepSeek-V4-Flash1 of 1 test
- Following instructionsDoing exactly what it is asked10DeepSeek-V4-Flash1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
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-V4-Flash pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+43.1points ahead
- Reasons across sets of long documentsAA-LCR+42.4points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+33points ahead
Where North-Mini-Code-1.0 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+8.5points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+1.2points ahead
Every test, side by side
All 21 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDeepSeek-V4-Flash
- Terminal-Bench 2.1DeepSeek-V4-Flash by 43.178.735.6+43.1
- LiveCodeBench v6DeepSeek-V4-Flash by 20.390.670.3+20.3
- SWE-bench ProDeepSeek-V4-Flash by 12.452.640.2+12.4
- SciCodeDeepSeek-V4-Flash by 11.550.338.8+11.5
- Terminal-Bench HardDeepSeek-V4-Flash by 4.535.631.1+4.5
- SWE-bench VerifiedNorth-Mini-Code-1.0 by 1.27980.2+1.2
AgentsDeepSeek-V4-Flash
- GDPValDeepSeek-V4-Flash by 46.946.90+46.9
- τ-Bench V3 · BankingDeepSeek-V4-Flash by 3339.46.4+33
- Terminal-Bench 4.0DeepSeek-V4-Flash by 11.612.10.5+11.6
ReasoningDeepSeek-V4-Flash
- Humanity's Last ExamDeepSeek-V4-Flash by 27.538.611.1+27.5
- CritPtDeepSeek-V4-Flash by 16.316.60.3+16.3
- GPQA DiamondDeepSeek-V4-Flash by 15.190.875.7+15.1
FactsEven
- AA-Omniscience · AccuracyDeepSeek-V4-Flash by 21.540.418.9+21.5
- AA-Omniscience · Non-hallucinationNorth-Mini-Code-1.0 by 8.58.316.8+8.5
Long documentsDeepSeek-V4-Flash
- AA-LCRDeepSeek-V4-Flash by 42.479.737.3+42.4
Following instructionsDeepSeek-V4-Flash
- IFBenchDeepSeek-V4-Flash by 21.679.257.6+21.6
Other results5 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.
- τ²-Bench Telecom (AA run)DeepSeek-V4-Flash by 57.69537.4+57.6
- AA-OmniscienceDeepSeek-V4-Flash by 34.3-14.3-48.6+34.3
- Artificial Analysis Coding IndexDeepSeek-V4-Flash by 32.669.136.5+32.6
- AA IntelligenceDeepSeek-V4-Flash by 24.434.39.9+24.4
- Terminal-Bench 2.0DeepSeek-V4-Flash by 1.856.955.1+1.8
Questions people ask
Which is better, DeepSeek-V4-Flash or North-Mini-Code-1.0?
DeepSeek-V4-Flash wins five of the six areas where both have results: coding, agents, reasoning, long documents and following instructions. North-Mini-Code-1.0 wins none. They are level on facts.
Which is better for coding?
DeepSeek-V4-Flash. It wins 5 of the 6 coding tests both models report; North-Mini-Code-1.0 wins 1.
How do you compare the two?
We use the 21 benchmark tests both models have published scores on. The verdict counts the 16 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 5 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.