Inkling vs Qwen3.5 397B A17B
Wins 6 of 8 areas
Coding · Agents · Reasoning · Facts · Math · Following instructions
Wins 1 of 8 areas
Images and charts
Inkling is the stronger all-rounder.Qwen3.5 397B A17B is cheaper and better at images and charts.
Both rank among the ten best models we track in following instructions.
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.
- CodingWriting and fixing software40Inkling4 of 4 tests
- AgentsCarrying out multi-step tasks on its own30Inkling3 of 3 tests
- FactsGetting facts right instead of making them up30Inkling3 of 3 tests
- ReasoningHard problems that need careful thinking21Inkling2 of 3 tests
- MathCompetition and research-level math21Inkling2 of 3 tests
- Following instructionsDoing exactly what it is asked10Inkling1 of 1 test
- Images and chartsUnderstanding pictures, charts and video01Qwen3.5 397B A17B1 of 1 test
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
Images and charts, 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.
Qwen3.5 397B A17B costs 17% 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 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+21.2points ahead
- Short factual questions, answered correctlySimpleQA Verified+17.9points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+15.7points ahead
Where Qwen3.5 397B A17B pulls ahead
- Harder college exam questions with imagesMMMU-Pro+3.8points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+2.1points ahead
- Harvard-MIT high-school math contest problemsHMMT Feb. 2026+1.6points 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
- Terminal-Bench 2.1Inkling by 3.855.151.3+3.8
- SciCodeInkling by 2.24744.8+2.2
- LMArena · WebDevInkling by 15 rating points14151400+15 rating
- SWE-bench VerifiedInkling by 1.277.676.4+1.2
AgentsInkling
- τ-Bench V3 · BankingInkling by 15.729.113.4+15.7
- GDPValInkling by 14.228.814.6+14.2
- Terminal-Bench 4.0Inkling by 110+1
ReasoningInkling
- CritPtInkling by 3.75.41.7+3.7
- Humanity's Last ExamInkling by 2.931.929+2.9
- GPQA DiamondQwen3.5 397B A17B by 2.187.289.3+2.1
FactsInkling
- AA-Omniscience · Non-hallucinationInkling by 21.232.311.1+21.2
- SimpleQA VerifiedInkling by 17.943.926+17.9
- AA-Omniscience · AccuracyInkling by 10.841.530.8+10.8
Images and chartsQwen3.5 397B A17B
- MMMU-ProQwen3.5 397B A17B by 3.873.577.3+3.8
MathInkling
- AIME 2026Inkling by 5.897.191.3+5.8
- FrontierMath Tiers 1-3 (v2)Inkling by 2.133.331.2+2.1
- HMMT Feb. 2026Qwen3.5 397B A17B by 1.686.387.9+1.6
Other results11 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-OmniscienceInkling by 32.72-30.8+32.7
- GDPval-AA v2Inkling by 276 rating points1238962+276 rating
- AA Agentic IndexInkling by 13.724.310.6+13.7
- τ³-Bench BankingInkling by 10.323.713.4+10.3
- AA IntelligenceInkling by 6.62518.4+6.6
- Toolathlon VerifiedInkling by 4.845.540.7+4.8
- Artificial Analysis Coding IndexInkling by 3.952.148.2+3.9
- SWEBench Pro PublicInkling by 3.454.350.9+3.4
- HLE (with tools)Qwen3.5 397B A17B by 2.34648.3+2.3
- BrowseComp (context management)Qwen3.5 397B A17B by 1.577.178.6+1.5
- StrongREJECTtie98.699.4tie
Questions people ask
Which is better, Inkling or Qwen3.5 397B A17B?
Inkling wins six of the eight areas we test: coding, agents, reasoning, facts, math and following instructions. Qwen3.5 397B A17B wins images and charts, and costs 17% less. They are level on long documents.
Which is better for coding?
Inkling. It wins 4 of the 4 coding tests both models report; Qwen3.5 397B A17B wins none.
Which is cheaper?
Inkling costs $1.00 per million input tokens and $4.05 per million output tokens; Qwen3.5 397B A17B costs $0.60 and $3.60. That makes Qwen3.5 397B A17B about 17% 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 19 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 11 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.