MiMo V2.6 Flash RL vs MiMo V2.6 Pro
Wins 0 of 6 areas
—
Wins 5 of 6 areas
Coding · Agents · Reasoning · Facts · Long documents
MiMo V2.6 Pro is the stronger all-rounder.MiMo V2.6 Flash RL is cheaper.
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.
- AgentsCarrying out multi-step tasks on its own03MiMo V2.6 Pro3 of 3 tests
- ReasoningHard problems that need careful thinking02MiMo V2.6 Pro2 of 2 tests
- FactsGetting facts right instead of making them up02MiMo V2.6 Pro2 of 2 tests
- CodingWriting and fixing software12MiMo V2.6 Pro2 of 3 tests
- Long documentsFinding answers in very long texts01MiMo V2.6 Pro1 of 1 test
- Images and chartsUnderstanding pictures, charts and video00Even0 each · 1 tie
Images and charts and long documents 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 V2.6 Flash RL costs 68% 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 MiMo V2.6 Pro pulls ahead
- Research-level physics problemsCritPt+14.6points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+14.3points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+13.8points ahead
Where MiMo V2.6 Flash RL pulls ahead
No clear win on a test scored out of 100.
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.
CodingMiMo V2.6 Pro
- SciCodeMiMo V2.6 Pro by 9.651.360.9+9.6
- Terminal-Bench 2.1MiMo V2.6 Pro by 2.387.689.9+2.3
- LMArena · WebDevMiMo V2.6 Flash RL by 14 rating points16381624+14 rating
AgentsMiMo V2.6 Pro
- Terminal-Bench 4.0MiMo V2.6 Pro by 12.122.734.8+12.1
- GDPValMiMo V2.6 Pro by 3.855.459.2+3.8
- OSWorld-VerifiedMiMo V2.6 Pro by 1.280.882+1.2
ReasoningMiMo V2.6 Pro
- CritPtMiMo V2.6 Pro by 14.61226.6+14.6
- Humanity's Last ExamMiMo V2.6 Pro by 14.335.149.4+14.3
FactsMiMo V2.6 Pro
- AA-Omniscience · Non-hallucinationMiMo V2.6 Pro by 13.845.659.4+13.8
- AA-Omniscience · AccuracyMiMo V2.6 Pro by 7.92734.9+7.9
Other results13 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.
- AA-OmniscienceMiMo V2.6 Pro by 21.1-12.78.4+21.1
- GDP (Surge AI)MiMo V2.6 Pro by 10.2919.2+10.2
- AA IntelligenceMiMo V2.6 Pro by 8.437.946.3+8.4
- Terminal-Bench 4.0MiMo V2.6 Pro by 6.128.834.9+6.1
- CritPt v2 HardMiMo V2.6 Pro by 4.529.534+4.5
- Agents' Last ExamMiMo V2.6 Pro by 427.631.6+4
- DeepSWE 1.1MiMo V2.6 Pro by 467.971.9+4
- Toolathlon VerifiedMiMo V2.6 Pro by 3.373.676.9+3.3
- HLE-DiamondMiMo V2.6 Pro by 1.513.715.2+1.5
- Terminal-Bench-Science 0.1MiMo V2.6 Pro by 1.44.35.7+1.4
- CyberGymMiMo V2.6 Flash RL by 1.195.194+1.1
- JobBenchtie61.262tie
- Program Benchtie2626.5tie
Questions people ask
Which is better, MiMo V2.6 Flash RL or MiMo V2.6 Pro?
MiMo V2.6 Pro wins five of the six areas where both have results: coding, agents, reasoning, facts and long documents. MiMo V2.6 Flash RL wins none, but costs 68% less. They are level on images and charts.
Which is better for coding?
MiMo V2.6 Pro. It wins 2 of the 3 coding tests both models report; MiMo V2.6 Flash RL wins 1.
Which is cheaper?
MiMo V2.6 Flash RL costs $0.14 per million input tokens and $0.28 per million output tokens; MiMo V2.6 Pro costs $0.43 and $0.87. That makes MiMo V2.6 Flash RL about 68% 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 12 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 13 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.