ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Qwen3.6 27B vs Claude Sonnet 4.5 vs GLM-5

Qwen3.6 27B comes out ahead, 64 to 57 and 56 on our weighted score, and it is the cheaper option too.

Price in rating: Included · Exclude price

  1. Our pick

    Alibaba (Qwen)

    Qwen3.6 27B

    Released Apr 22, 2026

    64/100
    • ECI146.8
    • Price$0.60 / $3.60
    • Context262K
  2. Anthropic

    Claude Sonnet 4.5

    Released Sep 29, 2025Deprecated

    57/100
    • ECI147.6
    • Price$3.00 / $15.00
    • Context200K
  3. Z.ai (Zhipu)

    GLM-5

    Released Feb 12, 2026

    56/100
    • ECI146.5
    • Price$1.00 / $3.20
    • Context205K
01 — Verdict

Qwen3.6 27B is our pick

Qwen3.6 27B is the better all-round choice, scoring 64/100 against Claude Sonnet 4.5 (57) and GLM-5 (56). It leads on price and context window. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

  • CapabilityClaude Sonnet 4.5Capabilities Index (ECI): Claude Sonnet 4.5 147.6 · Qwen3.6 27B 146.8 · GLM-5 146.5
  • Lowest priceQwen3.6 27BQwen3.6 27B $1.35 · GLM-5 $1.55 · Claude Sonnet 4.5 $6.00 per 1M tokens (3:1 blend)
  • Longest contextQwen3.6 27BQwen3.6 27B 262,144 · GLM-5 204,800 · Claude Sonnet 4.5 200,000 tokens
  • Widest inputsQwen3.6 27B and Claude Sonnet 4.5Qwen3.6 27B: Text, Images, Video · Claude Sonnet 4.5: Text, Images, PDFs · GLM-5: Text
  • Self-hostingQwen3.6 27B and GLM-5Publishes downloadable weights
How the score is built
MeasureWeightQwen3.6 27BClaude Sonnet 4.5GLM-5
CapabilityCapabilities Index (ECI)50%747574
Price25%421239
Inputs & features15%808035
Context window10%433939
Overall100%64/10057/10056/100
02 — Side by side

Every spec in one table

Highlighted cells lead their row. Dashes mean the data is not published.

Qwen3.6 27B vs Claude Sonnet 4.5 vs GLM-5 specifications side by side
SpecificationQwen3.6 27BAlibaba (Qwen)Claude Sonnet 4.5AnthropicGLM-5Z.ai (Zhipu)
Capability
Capabilities Index (ECI)146.8147.6 (best)146.5
ECI rank#84 of 158#78 of 158 (best)#85 of 158
GPQA DiamondGraduate-level science questions85.9%82.3%87.8% (best)
FrontierMath Tiers 1–3Research-level mathematics35.1% (best)23.9%—
OTIS Mock AIME 2024–2025Competition mathematics91.1% (best)77.8%80.0%
SWE-bench VerifiedFixing real GitHub issues—71.3%72.1% (best)
SimpleQA VerifiedShort factual questions—30.7%—
Price per million tokens
Input$0.60 (best)$3.00$1.00
Output$3.60$15.00$3.20 (best)
Cached input—$0.30$0.20 (best)
Blended (3:1)$1.35 (best)$6.00$1.55
Long-context rateSame rateSame rateSame rate
Price sourceOfficial Alibaba APIOfficial Anthropic APIOfficial Z.AI API
Limits
Context window262,144 tokens (best)200,000 tokens204,800 tokens
Max output65,536 tokens64,000 tokens131,072 tokens (best)
Inputs and features
TextYesYesYes
ImagesYesYesNo
PDFsNoYesNo
AudioNoNoNo
VideoYesNoNo
ReasoningYesYesYes
Tool callingYesYesYes
Structured outputYesYesNo
Availability
WeightsOpenProprietaryOpen
API model IDqwen3.6-27bclaude-sonnet-4-5glm-5
API providers2940 (best)37
ReleasedApr 22, 2026Sep 29, 2025Feb 12, 2026
Knowledge cutoff—Jul 31, 2025—
03 — Cost

What would a month cost?

Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.

  • Qwen3.6 27B$13.20
  • Claude Sonnet 4.5$60.00
  • GLM-5$16.40
04 — Questions

Which should you choose?

Which is better: Qwen3.6 27B, Claude Sonnet 4.5 or GLM-5?

Qwen3.6 27B is the better all-round choice, scoring 64/100 against Claude Sonnet 4.5 (57) and GLM-5 (56). It leads on price and context window. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

Which is cheaper, Qwen3.6 27B, Claude Sonnet 4.5 or GLM-5?

Qwen3.6 27B is cheaper at $0.60 input / $3.60 output per million tokens (official Alibaba API price). GLM-5 costs $1.00 input / $3.20 output per million tokens (official Z.AI API price); Claude Sonnet 4.5 costs $3.00 input / $15.00 output per million tokens (official Anthropic API price). At a typical mix of three input tokens to one output token, that is $1.35 per million tokens for Qwen3.6 27B versus $1.55 for GLM-5 (1.1× as much) and $6.00 for Claude Sonnet 4.5 (4.4× as much).

Which scores higher on benchmarks?

Claude Sonnet 4.5 scores higher on the Capabilities Index (ECI): Claude Sonnet 4.5 147.6 (#78 of 158), Qwen3.6 27B 146.8 (#84 of 158) and GLM-5 146.5 (#85 of 158). The confidence ranges of the top two overlap (146.1–149.0 vs 144.7–148.9), so treat the gap as small. On individual benchmarks: GPQA Diamond — GLM-5 87.8%, Qwen3.6 27B 85.9%, Claude Sonnet 4.5 82.3%; OTIS Mock AIME 2024–2025 — Qwen3.6 27B 91.1%, GLM-5 80.0%, Claude Sonnet 4.5 77.8%.

Which is better for coding?

There are no published SWE-bench Verified results for Qwen3.6 27B yet, so there is no like-for-like coding score. On overall capability, Claude Sonnet 4.5 leads, which tends to carry over to coding, but test on your own codebase. All three support tool calling for agent workflows.

Which has the bigger context window?

Qwen3.6 27B has the largest context window at 262,144 tokens, against 204,800 for GLM-5 and 200,000 for Claude Sonnet 4.5. Maximum output per response: Qwen3.6 27B up to 65,536, Claude Sonnet 4.5 up to 64,000, GLM-5 up to 131,072 tokens.

Which can read images, PDFs, audio or video?

Qwen3.6 27B accepts text, images and video; Claude Sonnet 4.5 accepts text, images and PDFs; GLM-5 accepts text. Qwen3.6 27B handles the widest range of inputs.

Are any of these open source?

Qwen3.6 27B and GLM-5 publishes its weights and can be self-hosted; Claude Sonnet 4.5 is proprietary.

Which is newer?

Qwen3.6 27B is the newest, released Apr 22, 2026. GLM-5 came out Feb 12, 2026; Claude Sonnet 4.5 came out Sep 29, 2025. Knowledge cutoff: Claude Sonnet 4.5 Jul 31, 2025.

How do you decide the winner?

This rating uses capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Nov 18, 2026.