DeepSeek-V4-Flash vs Qwen3.5 35B A3B
Wins 6 of 7 areas
Coding · Agents · Reasoning · Math · Long documents · Following instructions
Wins 0 of 7 areas
—
DeepSeek-V4-Flash is the stronger all-rounder.
Scores updated · 32 tests both models report · How we compare
Where each one wins
Tests won in each of the seven 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 software80DeepSeek-V4-Flash8 of 8 tests
- AgentsCarrying out multi-step tasks on its own50DeepSeek-V4-Flash5 of 5 tests
- ReasoningHard problems that need careful thinking30DeepSeek-V4-Flash3 of 3 tests
- MathCompetition and research-level math30DeepSeek-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.
DeepSeek-V4-Flash costs 22% less for the same work.
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
- Real work tasks from 44 professionsGDPVal+41.3points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+37.9points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+34.5points ahead
Where Qwen3.5 35B A3B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+6.3points ahead
Every test, side by side
All 32 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 37.978.740.8+37.9
- LMArena · WebDevDeepSeek-V4-Flash by 180 rating points14301250+180 rating
- LiveCodeBench v6DeepSeek-V4-Flash by 1690.674.6+16
- SWE-bench MultilingualDeepSeek-V4-Flash by 1373.360.3+13
- SciCodeDeepSeek-V4-Flash by 12.650.337.7+12.6
- SWE-bench VerifiedDeepSeek-V4-Flash by 9.87969.2+9.8
- Terminal-Bench HardDeepSeek-V4-Flash by 9.135.626.5+9.1
- SWE-bench ProDeepSeek-V4-Flash by 852.644.6+8
AgentsDeepSeek-V4-Flash
- GDPValDeepSeek-V4-Flash by 41.346.95.6+41.3
- τ-Bench V3 · BankingDeepSeek-V4-Flash by 34.539.44.9+34.5
- BrowseCompDeepSeek-V4-Flash by 12.273.261+12.2
- AA IT-Bench SREDeepSeek-V4-Flash by 1031.521.5+10
- MCP AtlasDeepSeek-V4-Flash by 6.66962.4+6.6
ReasoningDeepSeek-V4-Flash
- Humanity's Last ExamDeepSeek-V4-Flash by 17.638.621+17.6
- CritPtDeepSeek-V4-Flash by 15.716.60.9+15.7
- GPQA DiamondDeepSeek-V4-Flash by 6.390.884.5+6.3
FactsEven
- AA-Omniscience · AccuracyDeepSeek-V4-Flash by 20.340.420.1+20.3
- AA-Omniscience · Non-hallucinationQwen3.5 35B A3B by 6.38.314.6+6.3
MathDeepSeek-V4-Flash
- HMMT Feb. 2026DeepSeek-V4-Flash by 1394.881.8+13
- IMOAnswerBenchDeepSeek-V4-Flash by 11.688.476.8+11.6
- AIME 2026DeepSeek-V4-Flash by 2.595.893.3+2.5
Following instructionsDeepSeek-V4-Flash
- IFBenchDeepSeek-V4-Flash by 6.779.272.5+6.7
Other results9 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.
- AA-OmniscienceDeepSeek-V4-Flash by 33.8-14.3-48.1+33.8
- NL2RepoDeepSeek-V4-Flash by 33.754.220.5+33.7
- Artificial Analysis Coding IndexDeepSeek-V4-Flash by 32.169.137+32.1
- AA Agentic IndexDeepSeek-V4-Flash by 29.941.711.8+29.9
- Terminal-Bench 2.0DeepSeek-V4-Flash by 16.456.940.5+16.4
- AA IntelligenceDeepSeek-V4-Flash by 1534.319.3+15
- τ²-Bench Telecom (AA run)DeepSeek-V4-Flash by 5.89589.2+5.8
- GPQA (unspecified)DeepSeek-V4-Flash by 4.388.584.2+4.3
- MMLU-Protie86.285.3tie
Questions people ask
Which is better, DeepSeek-V4-Flash or Qwen3.5 35B A3B?
DeepSeek-V4-Flash wins six of the seven areas where both have results: coding, agents, reasoning, math, long documents and following instructions. Qwen3.5 35B A3B wins none. They are level on facts.
Which is better for coding?
DeepSeek-V4-Flash. It wins 8 of the 8 coding tests both models report; Qwen3.5 35B A3B wins none.
Which is cheaper?
DeepSeek-V4-Flash costs $0.44 per million input tokens and $1.32 per million output tokens; Qwen3.5 35B A3B costs $0.25 and $2.00. That makes DeepSeek-V4-Flash about 22% cheaper for the same work.
How do you compare the two?
We use the 32 benchmark tests both models have published scores on. The verdict counts the 23 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 9 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.