MiniMax M2.7 wins more areas, narrowly.Ling-3.0-flash-Fin is cheaper and better at agents.
Scores updated · 15 tests both models report · How we compare
Where each one wins
Tests won in each of the five 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 up02MiniMax M2.72 of 2 tests
- Long documentsFinding answers in very long texts01MiniMax M2.71 of 1 test
- AgentsCarrying out multi-step tasks on its own30Ling-3.0-flash-Fin3 of 4 tests · 1 tie
- CodingWriting and fixing software11Even1 each
- ReasoningHard problems that need careful thinking11Even1 each
Long documents rests on a single test.
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.
Ling-3.0-flash-Fin costs 80% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where MiniMax M2.7 pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+8.9points ahead
- Code for real scientific research problemsSciCode+7.7points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+7points ahead
Where Ling-3.0-flash-Fin pulls ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+28.7points ahead
- Long professional tasks in banking, consulting and lawAA ApexAgents+16.8points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+7.1points ahead
Every test, side by side
All 15 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- SciCodeMiniMax M2.7 by 7.742.450.1+7.7
- Terminal-Bench 2.1Ling-3.0-flash-Fin by 7.162.555.4+7.1
AgentsLing-3.0-flash-Fin
- τ-Bench V3 · BankingLing-3.0-flash-Fin by 28.738.69.9+28.7
- AA ApexAgentsLing-3.0-flash-Fin by 16.827.410.6+16.8
- GDPValLing-3.0-flash-Fin by 4.630.325.7+4.6
- Terminal-Bench 4.0tie00tie
ReasoningEven
- Humanity's Last ExamMiniMax M2.7 by 722.629.6+7
- CritPtLing-3.0-flash-Fin by 22.60.6+2
FactsMiniMax M2.7
- AA-Omniscience · AccuracyMiniMax M2.7 by 8.917.926.8+8.9
- AA-Omniscience · Non-hallucinationMiniMax M2.7 by 3.960.564.4+3.9
Other results4 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.
- τ³-Bench BankingLing-3.0-flash-Fin by 32.1418.9+32.1
- Finance Agent v2Ling-3.0-flash-Fin by 31.959.827.9+31.9
- AA-OmniscienceMiniMax M2.7 by 15.4-14.60.8+15.4
- AA Intelligencetie22.622.8tie
Questions people ask
Which is better, Ling-3.0-flash-Fin or MiniMax M2.7?
MiniMax M2.7 wins two of the five areas where both have results: facts and long documents. Ling-3.0-flash-Fin wins agents, and costs 80% less. They are level on coding and reasoning.
Which is better for coding?
Neither. They win 1 coding test each of the 2 both models report.
Which is cheaper?
Ling-3.0-flash-Fin costs $0.07 per million input tokens and $0.22 per million output tokens; MiniMax M2.7 costs $0.30 and $1.20. That makes Ling-3.0-flash-Fin about 80% cheaper for the same work.
How do you compare the two?
We use the 15 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 4 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.