Inkling-Small vs MiMo V2.5
Wins 7 of 8 areas
Coding · Agents · Reasoning · Facts · Math · Long documents · Following instructions
Wins 1 of 8 areas
Images and charts
Inkling-Small is the stronger all-rounder.MiMo V2.5 is cheaper and better at images and charts.
Scores updated · 30 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- AgentsCarrying out multi-step tasks on its own30Inkling-Small3 of 3 tests
- ReasoningHard problems that need careful thinking30Inkling-Small3 of 3 tests
- MathCompetition and research-level math20Inkling-Small2 of 2 tests
- CodingWriting and fixing software32Inkling-Small3 of 6 tests · 1 tie
- FactsGetting facts right instead of making them up21Inkling-Small2 of 3 tests
- Long documentsFinding answers in very long texts10Inkling-Small1 of 1 test
- Following instructionsDoing exactly what it is asked10Inkling-Small1 of 1 test
- Images and chartsUnderstanding pictures, charts and video03MiMo V2.53 of 3 tests
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.
MiMo V2.5 costs 78% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Inkling-Small pulls ahead
- Recent programming contest problemsLiveCodeBench v6+52.8points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+49.4points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+16.4points ahead
Where MiMo V2.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+31.1points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+8.6points ahead
- Reasoning about charts from research papersCharXiv (RQ)+3.6points ahead
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.
CodingInkling-Small
- LiveCodeBench v6Inkling-Small by 52.888.335.5+52.8
- SWE-bench VerifiedInkling-Small by 49.480.230.8+49.4
- Terminal-Bench 2.1MiMo V2.5 by 8.655.163.7+8.6
- SciCodeInkling-Small by 5.849.743.9+5.8
- LMArena · WebDevMiMo V2.5 by 30 rating points14081438+30 rating
- SWE-bench Protie55.956.1tie
AgentsInkling-Small
- τ-Bench V3 · BankingInkling-Small by 10.118.88.7+10.1
- GDPValInkling-Small by 6.331.224.9+6.3
- Terminal-Bench 4.0Inkling-Small by 110+1
ReasoningInkling-Small
- Humanity's Last ExamInkling-Small by 6.133.327.2+6.1
- CritPtInkling-Small by 4.68.33.7+4.6
- GPQA DiamondInkling-Small by 4.689.584.9+4.6
FactsInkling-Small
- AA-Omniscience · Non-hallucinationMiMo V2.5 by 31.13768.1+31.1
- AA-Omniscience · AccuracyInkling-Small by 16.433.216.8+16.4
- SimpleQA VerifiedInkling-Small by 4.520.616.1+4.5
Images and chartsMiMo V2.5
- CharXiv (RQ)MiMo V2.5 by 3.677.481+3.6
- MMMU-ProMiMo V2.5 by 1.47475.4+1.4
- LMArena · VisionMiMo V2.5 by 12 rating points12361248+12 rating
MathInkling-Small
- HMMT Feb. 2026Inkling-Small by 7.690.282.6+7.6
- AIME 2026Inkling-Small by 1.995.593.6+1.9
Following instructionsInkling-Small
- IFBenchInkling-Small by 15.182.267.1+15.1
Other results8 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.
- HLE (with tools)Inkling-Small by 7.847.840+7.8
- τ³-Bench BankingInkling-Small by 6.515.59+6.5
- Toolathlon VerifiedInkling-Small by 5.354.449.1+5.3
- AA IntelligenceInkling-Small by 3.925.721.8+3.9
- AA-Omnisciencetie-8.9-9.8tie
- StrongREJECTtie98.499.3tie
- GDPval-AA v2tie12691265tie
- SWEBench Pro Publictie55.956.1tie
Questions people ask
Which is better, Inkling-Small or MiMo V2.5?
Inkling-Small wins seven of the eight areas we test: coding, agents, reasoning, facts, math, long documents and following instructions. MiMo V2.5 wins images and charts, and costs 78% less.
Which is better for coding?
Inkling-Small. It wins 3 of the 6 coding tests both models report; MiMo V2.5 wins 2, and 1 is a tie.
Which is cheaper?
Inkling-Small costs $0.30 per million input tokens and $1.20 per million output tokens; MiMo V2.5 costs $0.11 and $0.22. That makes MiMo V2.5 about 78% 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 22 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 8 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.