Inkling vs Qwen3.8 2.4T A95B
Wins 0 of 5 areas
—
Wins 4 of 5 areas
Coding · Agents · Reasoning · Long documents
Qwen3.8 2.4T A95B is the stronger all-rounder.Inkling is cheaper.
Scores updated · 16 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.
- ReasoningHard problems that need careful thinking04Qwen3.8 2.4T A95B4 of 4 tests
- AgentsCarrying out multi-step tasks on its own03Qwen3.8 2.4T A95B3 of 3 tests
- CodingWriting and fixing software02Qwen3.8 2.4T A95B2 of 2 tests
- Long documentsFinding answers in very long texts01Qwen3.8 2.4T A95B1 of 1 test
- FactsGetting facts right instead of making them up11Even1 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.
Inkling costs 37% 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 2.4T A95B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+28.5points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+26.9points ahead
- Real work tasks from 44 professionsGDPVal+26.8points ahead
Where Inkling pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+10.3points ahead
Every test, side by side
All 16 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3.8 2.4T A95B
- Terminal-Bench 2.1Qwen3.8 2.4T A95B by 26.955.182+26.9
- SciCodeQwen3.8 2.4T A95B by 7.14754.1+7.1
AgentsQwen3.8 2.4T A95B
- GDPValQwen3.8 2.4T A95B by 26.828.855.6+26.8
- τ-Bench V3 · BankingQwen3.8 2.4T A95B by 2029.149.1+20
- Terminal-Bench 4.0Qwen3.8 2.4T A95B by 10.1111.1+10.1
ReasoningQwen3.8 2.4T A95B
- CritPtQwen3.8 2.4T A95B by 14.65.420+14.6
- SimpleBenchQwen3.8 2.4T A95B by 12.55062.5+12.5
- Humanity's Last ExamQwen3.8 2.4T A95B by 10.531.942.4+10.5
- GPQA DiamondQwen3.8 2.4T A95B by 6.387.293.5+6.3
FactsEven
- AA-Omniscience · Non-hallucinationQwen3.8 2.4T A95B by 28.532.360.8+28.5
- AA-Omniscience · AccuracyInkling by 10.341.531.3+10.3
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.
- AA Agentic IndexQwen3.8 2.4T A95B by 26.124.350.4+26.1
- Artificial Analysis Coding IndexQwen3.8 2.4T A95B by 19.852.171.9+19.8
- AA IntelligenceQwen3.8 2.4T A95B by 14.92539.9+14.9
- AA-OmniscienceQwen3.8 2.4T A95B by 2.324.3+2.3
Questions people ask
Which is better, Inkling or Qwen3.8 2.4T A95B?
Qwen3.8 2.4T A95B wins four of the five areas where both have results: coding, agents, reasoning and long documents. Inkling wins none, but costs 37% less. They are level on facts.
Which is better for coding?
Qwen3.8 2.4T A95B. It wins 2 of the 2 coding tests both models report; Inkling wins none.
Which is cheaper?
Inkling costs $1.00 per million input tokens and $4.05 per million output tokens; Qwen3.8 2.4T A95B costs $2.00 and $6.00. That makes Inkling about 37% cheaper for the same work.
How do you compare the two?
We use the 16 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 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.