Skip to content
aiqbench

Alibaba · model passport

Qwen: Qwen3.8 27B

qwen3-8-27b
  • Open weights
  • Apache 2.0 licence
  • 27.8B params
  • 1,000,000 context
  • Released 2026-08-05(HF repo created)
9/27axes covered
Overall · LMArena
1,438#89 of 396
Best percentile
82.0WebDev Arena
Cohort standing
62.7median 59.9

What the data says

Benchmarked on 9 of 27 tracked axes. Every figure below is the model's own published result on that board.

Uneven: 51.1 percentile points between its best and worst axis — 82 pct on WebDev Arena versus 30.9 pct on Agent Arena.

  • WebDev Arena82 pct1,594Verify →
  • Coding81.8 pct1,499Verify →
  • Hard prompts78.8 pct1,461Verify →
  • Overall77.8 pct1,438Verify →
  • Instruction following76 pct1,424Verify →
  • Vision Arena65.3 pct1,244Verify →
  • ARC-AGI-259.6 pct42.4%Verify →
  • Factuality Arena50 pct1,439Verify →
  • Agent Arena30.9 pct-0.034Verify →
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • GPQA DiamondNot benchmarked
  • AIME (math)Not benchmarked
  • FrontierMathNot benchmarked
  • SimpleQA VerifiedNot benchmarked
  • Humanity's Last ExamNot benchmarked
  • SimpleBenchNot benchmarked
  • LiveBenchNot benchmarked
  • BFCL v4 (function calling)Not benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not benchmarked
  • MMLU-Pro (Artificial Analysis)Not benchmarked
  • LiveCodeBench (Artificial Analysis)Not benchmarked
  • IFBench (Artificial Analysis)Not benchmarked
  • HHEM Hallucination (Vectara)Not benchmarked
  • SWE-bench VerifiedNot benchmarked
  • Terminal-BenchNot benchmarked

Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.

How to run it

Run it yourself

RepositoryQwen/Qwen3.8-27B
Parameters27.8B
ArchitectureQwen3_5ForConditionalGeneration
LicenceApache 2.0

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "Qwen/Qwen3.8-27B"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

Ollama

We don't sync Ollama tags, so we won't print a command we can't verify — a wrong one would pull the wrong model. Search the Ollama library →

Unsloth fine-tuning recipes are model-specific and we don't derive them here. Unsloth docs →

Use it hosted

API model IDqwen/qwen3.8-27b
Context length1,000,000 tokens
Max output235,929 tokens
Acceptstext, image, video
Input price$0.24 per 1M tokens
Cached input price$0.05 per 1M tokens
Output price$2.20 per 1M tokens

API parameters this endpoint supports

Frequency penaltyInclude reasoningLogit biasLogprobsMax tokensMin-pPresence penaltyReasoningReasoning effortRepetition penaltyResponse formatSeedStop sequencesStructured outputsTemperatureTool choiceTool callingTop-aTop-kTop logprobsTop-p

Example request

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen/qwen3.8-27b","messages":[{"role":"user","content":"Hello"}]}'

View on OpenRouter →

Identifiers, limits and prices are synced from the upstream source shown above. Any field we haven't synced is left out entirely rather than guessed.

  • qwen3.8-27b stands at 62.7 (raw mean 66.9) — measured on 9 of 27 axes.

    standingtemplateVerify →

  • qwen3.8-27b stands at 62.7 versus a cohort median of 59.9 across 191 models released within 180 days — above the peer median.

    cohorttemplateVerify →

Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →