DeepSeek-V4-Flash vs Qwen3.5 122B A10B
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 · 39 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 software80DeepSeek-V4-Flash8 of 8 tests
- AgentsCarrying out multi-step tasks on its own40DeepSeek-V4-Flash4 of 4 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.
DeepSeek-V4-Flash costs 51% 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
- Command-line tasks in a real terminalTerminal-Bench 2.1+31.1points ahead
- Real work tasks from 44 professionsGDPVal+31points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+24.1points ahead
Where Qwen3.5 122B A10B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+4.6points ahead
Every test, side by side
All 39 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 31.178.747.6+31.1
- LiveCodeBench v6DeepSeek-V4-Flash by 11.790.678.9+11.7
- SWE-bench MultilingualDeepSeek-V4-Flash by 11.373.362+11.3
- SciCodeDeepSeek-V4-Flash by 10.650.339.7+10.6
- SWE-bench ProDeepSeek-V4-Flash by 7.752.644.9+7.7
- LMArena · WebDevDeepSeek-V4-Flash by 71 rating points14311360+71 rating
- SWE-bench VerifiedDeepSeek-V4-Flash by 77972+7
- Terminal-Bench HardDeepSeek-V4-Flash by 4.535.631.1+4.5
AgentsDeepSeek-V4-Flash
- GDPValDeepSeek-V4-Flash by 3146.715.7+31
- τ-Bench V3 · BankingDeepSeek-V4-Flash by 24.139.415.3+24.1
- Terminal-Bench 4.0DeepSeek-V4-Flash by 12.112.10+12.1
- BrowseCompDeepSeek-V4-Flash by 9.473.263.8+9.4
ReasoningDeepSeek-V4-Flash
- CritPtDeepSeek-V4-Flash by 1616.60.6+16
- Humanity's Last ExamDeepSeek-V4-Flash by 13.438.625.2+13.4
- GPQA DiamondDeepSeek-V4-Flash by 5.190.885.7+5.1
FactsEven
- AA-Omniscience · AccuracyDeepSeek-V4-Flash by 1640.424.4+16
- AA-Omniscience · Non-hallucinationQwen3.5 122B A10B by 4.68.312.9+4.6
Long documentsDeepSeek-V4-Flash
- AA-LCRDeepSeek-V4-Flash by 3.479.776.3+3.4
Following instructionsDeepSeek-V4-Flash
- IFBenchDeepSeek-V4-Flash by 3.579.275.7+3.5
Other results20 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.
- DeepSWE [offline]DeepSeek-V4-Flash by 41.942.30.4+41.9
- GDPval-AA v2DeepSeek-V4-Flash by 328 rating points1189861+328 rating
- AA Agentic IndexDeepSeek-V4-Flash by 32.141.79.6+32.1
- AA-OmniscienceDeepSeek-V4-Flash by 27.2-14.3-41.5+27.2
- BrowseComp [offline]DeepSeek-V4-Flash by 258358+25
- Artificial Analysis Coding IndexDeepSeek-V4-Flash by 23.469.145.7+23.4
- tau2-bench airlineDeepSeek-V4-Flash by 2083.563.5+20
- AA IntelligenceDeepSeek-V4-Flash by 18.734.315.6+18.7
- tau2-bench retailDeepSeek-V4-Flash by 18.473.855.4+18.4
- HLE, text-onlyDeepSeek-V4-Flash by 11.635.824.2+11.6
- tau2-bench telecomDeepSeek-V4-Flash by 1198.287.2+11
- SciCode, subtaskDeepSeek-V4-Flash by 9.549.339.8+9.5
- τ³-Bench BankingDeepSeek-V4-Flash by 8.722.914.2+8.7
- Terminal-Bench 2.0DeepSeek-V4-Flash by 7.556.949.4+7.5
- WMT24++DeepSeek-V4-Flash by 7.285.578.3+7.2
- Terminal Bench 4 [offline]DeepSeek-V4-Flash by 3.53.50+3.5
- MMLU-ProXDeepSeek-V4-Flash by 3.385.582.2+3.3
- LMArena v3 proxyDeepSeek-V4-Flash by 2.849.847+2.8
- τ²-Bench Telecom (AA run)DeepSeek-V4-Flash by 1.49593.6+1.4
- MMLU-Protie86.286.7tie
Questions people ask
Which is better, DeepSeek-V4-Flash or Qwen3.5 122B A10B?
DeepSeek-V4-Flash wins five of the six areas where both have results: coding, agents, reasoning, long documents and following instructions. Qwen3.5 122B A10B 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 122B A10B wins none.
Which is cheaper?
DeepSeek-V4-Flash costs $0.44 per million input tokens and $1.32 per million output tokens; Qwen3.5 122B A10B costs $0.40 and $3.20. That makes DeepSeek-V4-Flash about 51% cheaper for the same work.
How do you compare the two?
We use the 39 benchmark tests both models have published scores on. The verdict counts the 19 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 20 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.