Inkling-Small vs Qwen3.8 Flash Next
Wins 0 of 7 areas
—
Wins 5 of 7 areas
Coding · Agents · Reasoning · Images and charts · Long documents
Qwen3.8 Flash Next is the stronger all-rounder.
Scores updated · 46 tests both models report · How we compare
Where each one wins
Tests won in each of the seven 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.
- CodingWriting and fixing software15Qwen3.8 Flash Next5 of 7 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own04Qwen3.8 Flash Next4 of 4 tests
- ReasoningHard problems that need careful thinking03Qwen3.8 Flash Next3 of 3 tests
- Images and chartsUnderstanding pictures, charts and video02Qwen3.8 Flash Next2 of 2 tests
- Long documentsFinding answers in very long texts01Qwen3.8 Flash Next1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
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.
Qwen3.8 Flash Next costs 59% 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 Qwen3.8 Flash Next pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+31points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+26.6points ahead
- Real work tasks from 44 professionsGDPVal+25.2points ahead
Where Inkling-Small pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+8.7points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+1.5points ahead
Every test, side by side
All 46 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3.8 Flash Next
- Terminal-Bench 2.1Qwen3.8 Flash Next by 3155.186.1+31
- LMArena · WebDevQwen3.8 Flash Next by 225 rating points14081633+225 rating
- SWE-bench MultilingualQwen3.8 Flash Next by 9.671.481+9.6
- SWE-bench ProQwen3.8 Flash Next by 6.655.962.5+6.6
- LiveCodeBench v6Qwen3.8 Flash Next by 3.688.391.9+3.6
- SWE-bench VerifiedInkling-Small by 1.580.278.7+1.5
- SciCodetie49.750.6tie
AgentsQwen3.8 Flash Next
- τ-Bench V3 · BankingQwen3.8 Flash Next by 26.618.845.4+26.6
- GDPValQwen3.8 Flash Next by 25.231.256.4+25.2
- Terminal-Bench 4.0Qwen3.8 Flash Next by 24.3125.3+24.3
- BrowseCompQwen3.8 Flash Next by 11.677.489+11.6
ReasoningQwen3.8 Flash Next
- Humanity's Last ExamQwen3.8 Flash Next by 4.733.338+4.7
- CritPtQwen3.8 Flash Next by 2.88.311.1+2.8
- GPQA DiamondQwen3.8 Flash Next by 2.889.592.3+2.8
FactsEven
- AA-Omniscience · Non-hallucinationQwen3.8 Flash Next by 17.73754.7+17.7
- AA-Omniscience · AccuracyInkling-Small by 8.733.224.5+8.7
Images and chartsQwen3.8 Flash Next
- CharXiv (RQ)Qwen3.8 Flash Next by 13.277.490.6+13.2
- MMMU-ProQwen3.8 Flash Next by 5.87479.8+5.8
Other results26 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.
- WebVoyager [offline]Qwen3.8 Flash Next by 85.15.390.4+85.1
- DeepSWE [offline]Qwen3.8 Flash Next by 37.61855.6+37.6
- GDPval-AA v2Qwen3.8 Flash Next by 323 rating points12381561+323 rating
- AA Agentic IndexQwen3.8 Flash Next by 31.42556.4+31.4
- τ³-Bench BankingQwen3.8 Flash Next by 2915.544.5+29
- BabyVisionQwen3.8 Flash Next by 22.124.746.8+22.1
- Artificial Analysis Coding IndexQwen3.8 Flash Next by 20.252.973.1+20.2
- Terminal Bench 4 [offline]Qwen3.8 Flash Next by 20.20.520.7+20.2
- ToolathlonQwen3.8 Flash Next by 19.154.473.5+19.1
- Toolathlon VerifiedQwen3.8 Flash Next by 19.154.473.5+19.1
- MMlongBench DocQwen3.8 Flash Next by 18.849.167.9+18.8
- LMArena v3 proxyInkling-Small by 18.453.635.2+18.4
- tau2-bench airlineQwen3.8 Flash Next by 18.162.780.8+18.1
- MATH-VisionQwen3.8 Flash Next by 16.978.895.7+16.9
- OCRBench v2, ZHQwen3.8 Flash Next by 15.952.468.3+15.9
- BrowseComp [offline]Qwen3.8 Flash Next by 157489+15
- AA IntelligenceQwen3.8 Flash Next by 14.125.739.8+14.1
- OCRBench v2_enQwen3.8 Flash Next by 13.754.267.9+13.7
- HLE, full setQwen3.8 Flash Next by 7.231.638.8+7.2
- HLE, text-onlyQwen3.8 Flash Next by 6.333.339.6+6.3
- SciCode, subtaskInkling-Small by 1.748.346.6+1.7
- WMT24++Qwen3.8 Flash Next by 1.684.686.2+1.6
- MMLU-ProXInkling-Small by 18483+1
- AA-Omnisciencetie-8.9-9.7tie
- tau2-bench telecomtie90.190.4tie
- tau2-bench retailtie71.871.6tie
Questions people ask
Which is better, Inkling-Small or Qwen3.8 Flash Next?
Qwen3.8 Flash Next wins five of the seven areas where both have results: coding, agents, reasoning, images and charts and long documents. Inkling-Small wins none. They are level on facts and following instructions.
Which is better for coding?
Qwen3.8 Flash Next. It wins 5 of the 7 coding tests both models report; Inkling-Small wins 1, and 1 is a tie.
Which is cheaper?
Inkling-Small costs $0.30 per million input tokens and $1.20 per million output tokens; Qwen3.8 Flash Next costs $0.15 and $0.47. That makes Qwen3.8 Flash Next about 59% cheaper for the same work.
How do you compare the two?
We use the 46 benchmark tests both models have published scores on. The verdict counts the 20 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 26 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.