Head to head

Qwen3.8 27B vs Qwen3.6 27B

Qwen3.8 27B scores higher on 5 of the 5 benchmarks both models publish. The widest gap is on Terminal-Bench, where Qwen3.8 27B leads by 13.7 points.

Qwen3.8 27BAlibaba Qwen · OpenLLM Buddy catalog83.2compositeQwen3.6 27BAlibaba Qwen · OpenLLM Buddy catalog85.8composite

Shared benchmarks

Raw published scores; the higher score on each row is highlighted. LMArena Elo is a human-preference vote and is not counted as a win.

Should you pick Qwen3.8 27B or Qwen3.6 27B?

Pick Qwen3.8 27B if…

  • You care most about Humanity's Last Exam, LiveCodeBench, SWE-bench Pro, where it scores at least 2 points higher.
  • You want it running in minutes on a dedicated GPU through OpenLLM Buddy, in an uncensored build.

Pick Qwen3.6 27B if…

  • You need fast responses: about 57 tokens per second through APIs, against 41.
  • You want it running in minutes on a dedicated GPU through OpenLLM Buddy.

Hardware and cost to run

Qwen3.8 27B

At 4-bit it needs about 19 GB, which fits a single RTX 4090, so it can run on a home workstation with llama.cpp, Ollama, or vLLM. Full BF16 precision needs about 62 GB (1× H100 (80 GB)).

On OpenLLM Buddy: Qwen3.8-27B-Uncensored-FP8 runs on H100 80GB as an FP8 build, with an OpenAI-compatible API and no per-token fees. Prices start at $53 USD for 11 hours.

Deploy Qwen3.8 27B
Full GPU memory table for Qwen3.8 27B

Qwen3.6 27B

At 4-bit it needs about 19 GB, which fits a single RTX 4090, so it can run on a home workstation with llama.cpp, Ollama, or vLLM. Full BF16 precision needs about 62 GB (1× H100 (80 GB)).

On OpenLLM Buddy: Qwen3.6 27B A4B, with an OpenAI-compatible API and no per-token fees. Prices start at $17 USD for 11 Hours.

Deploy Qwen3.6 27B
Full GPU memory table for Qwen3.6 27B

Category profile

Category index uses every test each model reports, not only the shared ones.

ReasoningCodingAgenticKnowledge
Qwen3.8 27BQwen3.6 27B
Reasoning70.380.9
Coding81.977.3
Agentic——
Knowledge97.399.3

Specs side by side

Qwen3.8 27BQwen3.6 27B
LabAlibaba QwenAlibaba Qwen
WeightsOpenOpen
Released2026-08-142026-04
Parameters27B dense27B dense
Context262K (up to 1M)262K (up to 1M)
LicenseApache-2.0Apache-2.0
4-bit memory19 GB · RTX 4090 (24 GB)19 GB · RTX 4090 (24 GB)
API price$0.425 in / $2.55 out per 1M tokens$0.450 in / $2.70 out per 1M tokens
API speed41 tok/s57 tok/s
One-GPU speed (est.)≈ 52–69 tok/s on H100 80GB (FP8)≈ 28–39 tok/s on RTX 4090 (4-bit)

Related comparisons

Sources

ModelTestScoreSettingSource
Qwen3.8 27BGPQA Diamond89.2—Qwen3.8-27B model card (Hugging Face)
Qwen3.6 27BGPQA Diamond87.8—Qwen3.6-27B model card (Hugging Face)
Qwen3.8 27BHumanity's Last Exam30.8—Qwen3.8-27B model card (Hugging Face)
Qwen3.6 27BHumanity's Last Exam24.0—Qwen3.6-27B model card (Hugging Face)
Qwen3.8 27BLiveCodeBench90.3LiveCodeBench v6Qwen3.8-27B model card (Hugging Face)
Qwen3.6 27BLiveCodeBench83.9LiveCodeBench v6Qwen3.6-27B model card (Hugging Face)
Qwen3.8 27BSWE-bench Pro61.7—Qwen3.8-27B model card (Hugging Face)
Qwen3.6 27BSWE-bench Pro53.5—Qwen3.6-27B model card (Hugging Face)
Qwen3.8 27BTerminal-Bench73.0Terminal Bench 2.1 (Terminus harness)Qwen3.8-27B model card (Hugging Face)
Qwen3.6 27BTerminal-Bench59.3Terminal-Bench 2.0Qwen3.6-27B model card (Hugging Face)

Scores are published results from model cards, lab reports, and public leaderboards, each linked to its source. Settings such as thinking mode or test version differ between sources and are shown next to each score. OpenLLM Buddy catalog models are uncensored fine-tunes; their scores are those of the official base model. Data collected 2026-10-30. How the index works