ResearchLLM Verifications Leaderboard

Data as of · 78 models, 80 questions · reviewed by Suhas Bhairav · methodology

Head to head · tested 27 Oct 2026

Qwen3.6 35B A3B vs Qwen3.8 2.4T A95B: accuracy, speed and price

Qwen3.8 2.4T A95B is more accurate; Qwen3.6 35B A3B is cheaper.

QwenQwen3.6 35B A3B37.5% score · #59$0.0037 per question135 tok/sQwenQwen3.8 2.4T A95B57.5% score · #43$0.015 per question44 tok/s

Verdict

Built from the measurements below

  • Qwen3.8 2.4T A95B is more accurate: 57.5% against 37.5% on the same 80 questions, a 20.0-point gap that is statistically significant.
  • Qwen3.6 35B A3B is 4.1× cheaper per question as billed ($0.0037 vs $0.015), reasoning tokens included.
  • Qwen3.6 35B A3B generates text 3.0× faster: 135 tokens per second against 44.
  • Qwen3.6 35B A3B starts answering sooner: 0.4 s to the first token, against 3.1 s for Qwen3.8 2.4T A95B.
  • Qwen3.8 2.4T A95B accepts more context: 1M tokens against 262k.

Which should you choose?

Choose Qwen3.6 35B A3B for the lower cost per correct answer ($0.0099 against $0.026), faster streaming, a quicker first token and more hosts to choose from on OpenRouter (10 against 5). Choose Qwen3.8 2.4T A95B for the most accurate answers of the two and a 1M-token context window. The widest subject gap is in science: Qwen3.8 2.4T A95B answered 21 of 30 correctly, Qwen3.6 35B A3B 9 of 30. With so few questions per subject, treat this as a lead to test on your own prompts, not a result.

Accuracy gap -20.0 points (z = -2.53). A winner is called only when z is at least 1.96, the usual 95% threshold. Speed, latency and price need a 10% margin.

Head to head

Same questions, same prompt, same API

MeasureQwen3.6 35B A3BQwen3.8 2.4T A95BEdge
Verification scoreHigher is better37.5%57.5%Qwen3.8 2.4T A95B
Cost per questionLower is better$0.0037$0.015Qwen3.6 35B A3B
Output speedHigher is better13544Qwen3.6 35B A3B
Time to first tokenLower is better0.4 s3.1 sQwen3.6 35B A3B
End-to-end response timeLower is better5.6 s14.8 sQwen3.6 35B A3B
Tokens per questionLower is better3,5632,408Qwen3.8 2.4T A95B
Context windowHigher is better262k1MQwen3.8 2.4T A95B

Side by side

Bars are scaled to the larger of the two values. The winner on each measure is marked in bold.

Every measure

Qwen3.6 35B A3B first, Qwen3.8 2.4T A95B second

Verification scorehigher is better
Qwen3.6 35B A3B37.5%
Qwen3.8 2.4T A95B57.5%
Cost per questionlower is better
Qwen3.6 35B A3B$0.0037
Qwen3.8 2.4T A95B$0.015
Output speedhigher is better
Qwen3.6 35B A3B135
Qwen3.8 2.4T A95B44
Time to first tokenlower is better
Qwen3.6 35B A3B0.4 s
Qwen3.8 2.4T A95B3.1 s
End-to-end response timelower is better
Qwen3.6 35B A3B5.6 s
Qwen3.8 2.4T A95B14.8 s
Tokens per questionlower is better
Qwen3.6 35B A3B3,563
Qwen3.8 2.4T A95B2,408
Context windowhigher is better
Qwen3.6 35B A3B262k
Qwen3.8 2.4T A95B1M

Score vs. cost per question

Both models in color, every other model in grey

Most attractive quadrant0255075100$0.00001$0.0001$0.001$0.01$0.10Cost per question (USD, log) →Verification score (%) →Qwen3.8 2.4T A95BCommand AReka Flash 3Gemma 4 26B A4BGemma 4 31BQwen3.5 397B A17BHermes 4 405BMuse WonderGrok 4.7o3gpt-oss-20bMinistral 3 3B 2512Nova Micro 1.0Phi 4Granite 4.2 8BHy3GLM 5.3 FlashSolar Pro 4Trinity Large ThinkingMiniMax M3Mistral Large 4Claude Opus 5.5GPT-6.1 SolGemini 4 ArgonGPT-5.4GPT-6 Lunagpt-oss-120bLlama 4 Maverick

Specifications and pricing

From OpenRouter's catalogue, plus the leaderboard's test dates

SpecificationQwen3.6 35B A3BQwen3.8 2.4T A95B
CreatorQwenQwen
Released27 Apr 202612 Aug 2026
WeightsOpen weightsOpen weights
ReasoningYesYes
Inputtext, image, videotext
Outputtexttext
Context window262k tokens1M tokens
Knowledge cutoff——
List price in / out per 1M$0.1 / $0.95$2 / $6
Blended price per 1M (3:1)$0.31$3.00
Cheapest hostDarkbloom · $0.05 / $0.7Novita · $2 / $6
Hosts on OpenRouter105
Last tested26 Oct 202627 Oct 2026

Compare any two models →

Qwen3.6 35B A3B vs Qwen3.8 2.4T A95B: frequently asked questions

Is Qwen3.6 35B A3B better than Qwen3.8 2.4T A95B?
Qwen3.8 2.4T A95B is more accurate: it answered 46 of 80 questions correctly (57.5%) against 30 of 80 (37.5%) for Qwen3.6 35B A3B. Qwen3.8 2.4T A95B is more accurate; Qwen3.6 35B A3B is cheaper.
Which is cheaper, Qwen3.6 35B A3B or Qwen3.8 2.4T A95B?
Qwen3.6 35B A3B is cheaper: $0.0037 per question as billed, against $0.015. List prices: Qwen3.6 35B A3B $0.1 in / $0.95 out, Qwen3.8 2.4T A95B $2 in / $6 out per million tokens.
Which is faster, Qwen3.6 35B A3B or Qwen3.8 2.4T A95B?
Qwen3.6 35B A3B is faster at 135 output tokens per second, against 44. Time to first token: 0.4 s for Qwen3.6 35B A3B, 3.1 s for Qwen3.8 2.4T A95B.
Which has the larger context window, Qwen3.6 35B A3B or Qwen3.8 2.4T A95B?
Qwen3.8 2.4T A95B, with 1M tokens against 262k.