Kimi K2 Instruct vs Seed Oss 36B Instruct
Wins 2 of 6 areas
Agents · Facts
Wins 2 of 6 areas
Coding · Long documents
The two are evenly matched.Kimi K2 Instruct is better at facts and agents; Seed Oss 36B Instruct at coding and long documents.
Scores updated · 25 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.
- FactsGetting facts right instead of making them up20Kimi K2 Instruct2 of 2 tests
- AgentsCarrying out multi-step tasks on its own10Kimi K2 Instruct1 of 1 test
- CodingWriting and fixing software12Seed Oss 36B Instruct2 of 3 tests
- Long documentsFinding answers in very long texts01Seed Oss 36B Instruct1 of 1 test
- ReasoningHard problems that need careful thinking11Even1 each · 1 tie
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
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.
Seed Oss 36B Instruct costs 75% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Kimi K2 Instruct pulls ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+16.7points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+16.7points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+7.2points ahead
Where Seed Oss 36B Instruct pulls ahead
- Recent programming contest problemsLiveCodeBench v6+13.7points ahead
- Reasons across sets of long documentsAA-LCR+8.3points ahead
- Code for real scientific research problemsSciCode+5.8points ahead
Every test, side by side
All 25 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingSeed Oss 36B Instruct
- Terminal-Bench HardKimi K2 Instruct by 16.723.56.8+16.7
- LiveCodeBench v6Seed Oss 36B Instruct by 13.753.767.4+13.7
- SciCodeSeed Oss 36B Instruct by 5.830.736.5+5.8
ReasoningEven
- GPQA DiamondKimi K2 Instruct by 4.176.772.6+4.1
- Humanity's Last ExamSeed Oss 36B Instruct by 3.56.49.9+3.5
- CritPttie00tie
FactsKimi K2 Instruct
- AA-Omniscience · Non-hallucinationKimi K2 Instruct by 16.730.313.6+16.7
- AA-Omniscience · AccuracyKimi K2 Instruct by 7.225.418.2+7.2
Long documentsSeed Oss 36B Instruct
- AA-LCRSeed Oss 36B Instruct by 8.35361.3+8.3
Other results14 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.
- AIME 2025Seed Oss 36B Instruct by 35.249.584.7+35.2
- AA-OmniscienceKimi K2 Instruct by 25.9-26.6-52.5+25.9
- τ²-Bench Telecom (AA run)Kimi K2 Instruct by 2473.449.4+24
- AIME 2024Seed Oss 36B Instruct by 22.169.691.7+22.1
- SimpleQAKimi K2 Instruct by 21.3319.7+21.3
- Multi-SWE-BenchKimi K2 Instruct by 16.533.517+16.5
- AA Agentic IndexKimi K2 Instruct by 1037.727.7+10
- Artificial Analysis Coding IndexKimi K2 Instruct by 9.225.916.7+9.2
- IFEvalKimi K2 Instruct by 489.885.8+4
- AA IntelligenceKimi K2 Instruct by 3.215.312.1+3.2
- MMLUKimi K2 Instruct by 2.189.587.4+2.1
- MMLU-ProSeed Oss 36B Instruct by 1.681.182.7+1.6
- AIR-Bench 2024Seed Oss 36B Instruct by 1.574.175.6+1.5
- SuperGPQAKimi K2 Instruct by 1.557.255.7+1.5
Questions people ask
Which is better, Kimi K2 Instruct or Seed Oss 36B Instruct?
Kimi K2 Instruct and Seed Oss 36B Instruct each win two of the six areas where both have results. Kimi K2 Instruct wins agents and facts; Seed Oss 36B Instruct wins coding and long documents. Seed Oss 36B Instruct costs 75% less. They are level on reasoning and following instructions.
Which is better for coding?
Seed Oss 36B Instruct. It wins 2 of the 3 coding tests both models report; Kimi K2 Instruct wins 1.
Which is cheaper?
Kimi K2 Instruct costs $0.60 per million input tokens and $2.50 per million output tokens; Seed Oss 36B Instruct costs $0.21 and $0.57. That makes Seed Oss 36B Instruct about 75% cheaper for the same work.
How do you compare the two?
We use the 25 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 14 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.