DeepSeek-V3.2-Speciale vs Kimi K2.5
Wins 0 of 6 areas
—
Wins 4 of 6 areas
Coding · Agents · Long documents · Following instructions
Kimi K2.5 is the stronger all-rounder.DeepSeek-V3.2-Speciale is cheaper.
Scores updated · 22 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 software02Kimi K2.52 of 3 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own01Kimi K2.51 of 1 test
- Long documentsFinding answers in very long texts01Kimi K2.51 of 1 test
- Following instructionsDoing exactly what it is asked01Kimi K2.51 of 1 test
- ReasoningHard problems that need careful thinking11Even1 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.
DeepSeek-V3.2-Speciale costs 80% 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 Kimi K2.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+23.5points ahead
- Reasons across sets of long documentsAA-LCR+8points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+6.3points ahead
Where DeepSeek-V3.2-Speciale pulls ahead
- Research-level physics problemsCritPt+4.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+2.6points ahead
Every test, side by side
All 22 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingKimi K2.5
- SciCodeKimi K2.5 by 54449+5
- SWE-bench VerifiedKimi K2.5 by 3.773.176.8+3.7
- Terminal-Bench Hardtie34.834.8tie
ReasoningEven
- CritPtDeepSeek-V3.2-Speciale by 4.37.43.1+4.3
- Humanity's Last ExamKimi K2.5 by 228.730.7+2
- GPQA Diamondtie87.187.9tie
FactsEven
- AA-Omniscience · Non-hallucinationKimi K2.5 by 23.510.834.3+23.5
- AA-Omniscience · AccuracyDeepSeek-V3.2-Speciale by 2.637.835.2+2.6
Following instructionsKimi K2.5
- IFBenchKimi K2.5 by 6.363.970.2+6.3
Other results11 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)Kimi K2.5 by 95.9095.9+95.9
- AA Agentic IndexKimi K2.5 by 21.7021.7+21.7
- AA-OmniscienceKimi K2.5 by 10.4-17.7-7.3+10.4
- AA IntelligenceKimi K2.5 by 914.523.5+9
- Artificial Analysis Coding IndexKimi K2.5 by 8.937.946.8+8.9
- ToolathlonDeepSeek-V3.2-Speciale by 7.435.227.8+7.4
- Terminal-Bench 2.0Kimi K2.5 by 4.446.450.8+4.4
- HMMT Nov. 2025DeepSeek-V3.2-Speciale by 4.193.389.2+4.1
- HMMT 2025DeepSeek-V3.2-Speciale by 3.899.295.4+3.8
- HMMT Feb. 2025DeepSeek-V3.2-Speciale by 2.197.595.4+2.1
- AIME 2025tie9696.1tie
Questions people ask
Which is better, DeepSeek-V3.2-Speciale or Kimi K2.5?
Kimi K2.5 wins four of the six areas where both have results: coding, agents, long documents and following instructions. DeepSeek-V3.2-Speciale wins none, but costs 80% less. They are level on reasoning and facts.
Which is better for coding?
Kimi K2.5. It wins 2 of the 3 coding tests both models report; DeepSeek-V3.2-Speciale wins none, and 1 is a tie.
Which is cheaper?
DeepSeek-V3.2-Speciale costs $0.29 per million input tokens and $0.43 per million output tokens; Kimi K2.5 costs $0.60 and $3.00. That makes DeepSeek-V3.2-Speciale about 80% cheaper for the same work.
How do you compare the two?
We use the 22 benchmark tests both models have published scores on. The verdict counts the 11 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 11 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.