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
- Our pick
Alibaba (Qwen)
Qwen3.6 27B
64/100- ECI146.8
- Price$0.60 / $3.60
- Context262K
Anthropic
Claude Sonnet 4.5
57/100- ECI147.6
- Price$3.00 / $15.00
- Context200K
Z.ai (Zhipu)
GLM-5
56/100- ECI146.5
- Price$1.00 / $3.20
- Context205K
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
| Measure | Weight | Qwen3.6 27B | Claude Sonnet 4.5 | GLM-5 |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 50% | 74 | 75 | 74 |
| Price | 25% | 42 | 12 | 39 |
| Inputs & features | 15% | 80 | 80 | 35 |
| Context window | 10% | 43 | 39 | 39 |
| Overall | 100% | 64/100 | 57/100 | 56/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | |||
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 146.8 | 147.6 (best) | 146.5 |
| ECI rank | #84 of 158 | #78 of 158 (best) | #85 of 158 |
| GPQA DiamondGraduate-level science questions | 85.9% | 82.3% | 87.8% (best) |
| FrontierMath Tiers 1–3Research-level mathematics | 35.1% (best) | 23.9% | — |
| OTIS Mock AIME 2024–2025Competition mathematics | 91.1% (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 rate | Same rate | Same rate | Same rate |
| Price source | Official Alibaba API | Official Anthropic API | Official Z.AI API |
| Limits | |||
| Context window | 262,144 tokens (best) | 200,000 tokens | 204,800 tokens |
| Max output | 65,536 tokens | 64,000 tokens | 131,072 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | Yes | No |
| PDFs | No | Yes | No |
| Audio | No | No | No |
| Video | Yes | No | No |
| Reasoning | Yes | Yes | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | Yes | No |
| Availability | |||
| Weights | Open | Proprietary | Open |
| API model ID | qwen3.6-27b | claude-sonnet-4-5 | glm-5 |
| API providers | 29 | 40 (best) | 37 |
| Released | Apr 22, 2026 | Sep 29, 2025 | Feb 12, 2026 |
| Knowledge cutoff | — | Jul 31, 2025 | — |
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
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.