Granite 3.3 8B Instruct vs Kimi K2 Instruct
Wins 0 of 6 areas
—
Wins 6 of 6 areas
Coding · Agents · Reasoning · Facts · Long documents · Following instructions
Kimi K2 Instruct is the stronger all-rounder.Granite 3.3 8B Instruct is cheaper.
Scores updated · 30 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.
- ReasoningHard problems that need careful thinking02Kimi K2 Instruct2 of 3 tests · 1 tie
- CodingWriting and fixing software02Kimi K2 Instruct2 of 2 tests
- FactsGetting facts right instead of making them up12Kimi K2 Instruct2 of 3 tests
- AgentsCarrying out multi-step tasks on its own01Kimi K2 Instruct1 of 1 test
- Long documentsFinding answers in very long texts01Kimi K2 Instruct1 of 1 test
- Following instructionsDoing exactly what it is asked01Kimi K2 Instruct1 of 1 test
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.
Granite 3.3 8B Instruct costs 91% 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 Instruct pulls ahead
- Reasons across sets of long documentsAA-LCR+49.7points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+42.9points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+26.5points ahead
Where Granite 3.3 8B Instruct pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 30 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingKimi K2 Instruct
- Terminal-Bench HardKimi K2 Instruct by 23.5023.5+23.5
- SciCodeKimi K2 Instruct by 20.610.130.7+20.6
ReasoningKimi K2 Instruct
- GPQA DiamondKimi K2 Instruct by 42.933.876.7+42.9
- Humanity's Last ExamKimi K2 Instruct by 2.24.26.4+2.2
- CritPttie00tie
FactsKimi K2 Instruct
- AA-Omniscience · Non-hallucinationKimi K2 Instruct by 26.53.830.3+26.5
- AA-Omniscience · AccuracyKimi K2 Instruct by 15.79.725.4+15.7
- Vectara HHEM hallucination ratelower is betterGranite 3.3 8B Instruct by 7.310.617.9+7.3
Following instructionsKimi K2 Instruct
- IFBenchKimi K2 Instruct by 19.322.441.7+19.3
Other results19 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 Instruct by 62.910.573.4+62.9
- AIME 2024Kimi K2 Instruct by 61.58.169.6+61.5
- AA-OmniscienceKimi K2 Instruct by 50.7-77.3-26.6+50.7
- AA Agentic IndexKimi K2 Instruct by 34.23.537.7+34.2
- MATH-500 (EM)Kimi K2 Instruct by 28.46997.4+28.4
- MMLUKimi K2 Instruct by 2465.589.5+24
- Artificial Analysis Coding IndexKimi K2 Instruct by 22.53.425.9+22.5
- GSM8KKimi K2 Instruct by 16.480.997.3+16.4
- HarmBenchKimi K2 Instruct by 16.281.297.4+16.2
- IFEvalKimi K2 Instruct by 1574.889.8+15
- AA IntelligenceKimi K2 Instruct by 10.44.915.3+10.4
- AIR-Bench 2024Granite 3.3 8B Instruct by 8.482.574.1+8.4
- XSTestKimi K2 Instruct by 7.590.798.2+7.5
- vectara_factual_consistencyGranite 3.3 8B Instruct by 7.389.482.1+7.3
- vectara_avg_summary_lengthGranite 3.3 8B Instruct by 72.2 rating points131.459.2+72.2 rating
- HumanEvalKimi K2 Instruct by 3.689.793.3+3.6
- simple_safety_testsKimi K2 Instruct by 2.597.5100+2.5
- vectara_answer_rateGranite 3.3 8B Instruct by 1.410098.6+1.4
- anthropic_red_teamtie99.199.3tie
Questions people ask
Which is better, Granite 3.3 8B Instruct or Kimi K2 Instruct?
Kimi K2 Instruct wins all six areas where both have results: coding, agents, reasoning, facts, long documents and following instructions. Granite 3.3 8B Instruct wins none, but costs 91% less.
Which is better for coding?
Kimi K2 Instruct. It wins 2 of the 2 coding tests both models report; Granite 3.3 8B Instruct wins none.
Which is cheaper?
Granite 3.3 8B Instruct costs $0.03 per million input tokens and $0.25 per million output tokens; Kimi K2 Instruct costs $0.60 and $2.50. That makes Granite 3.3 8B Instruct about 91% cheaper for the same work.
How do you compare the two?
We use the 30 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 19 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.