DeepSeek-R1 vs Qwen3 32B vs Claude Sonnet 3.5 v2
Too close to call on our weighted score (DeepSeek-R1 51, Qwen3 32B 51, Claude Sonnet 3.5 v2 44). The right pick depends on what you value most.
DeepSeekDeepSeek-R1
51/100- ECI139.0
- Price$0.70 / $2.60
- Context128K
Alibaba (Qwen)Qwen3 32B
51/100- ECI138.5
- Price$0.70 / $2.80
- Context131K
AnthropicClaude Sonnet 3.5 v2
44/100- ECI133.5
- Price$3.00 / $15.00
- Context200K
Too close to call
It is close. Our weighted score puts them within a point (DeepSeek-R1 51/100, Qwen3 32B 51/100, Claude Sonnet 3.5 v2 44/100), so choose by what matters most for your work: DeepSeek-R1 for raw capability and Claude Sonnet 3.5 v2 for long inputs. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
- CapabilityDeepSeek-R1Capabilities Index (ECI): DeepSeek-R1 139.0 · Qwen3 32B 138.5 · Claude Sonnet 3.5 v2 133.5
- Lowest priceDeepSeek-R1DeepSeek-R1 $1.18 · Qwen3 32B $1.23 · Claude Sonnet 3.5 v2 $6.00 per 1M tokens (3:1 blend)
- Longest contextClaude Sonnet 3.5 v2Claude Sonnet 3.5 v2 200,000 · Qwen3 32B 131,072 · DeepSeek-R1 128,000 tokens
- Widest inputsClaude Sonnet 3.5 v2DeepSeek-R1: Text · Qwen3 32B: Text · Claude Sonnet 3.5 v2: Text, Images, PDFs
- Self-hostingDeepSeek-R1 and Qwen3 32BPublishes downloadable weights
| Measure | Weight | DeepSeek-R1 | Qwen3 32B | Claude Sonnet 3.5 v2 |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 50% | 64 | 64 | 57 |
| Price | 25% | 47 | 46 | 13 |
| Inputs & features | 15% | 35 | 35 | 60 |
| Context window | 10% | 24 | 24 | 32 |
| Overall | 100% | 51/100 | 51/100 | 44/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | DeepSeek-R1DeepSeek | Qwen3 32BAlibaba (Qwen) | Claude Sonnet 3.5 v2Anthropic |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 139.0 (best) | 138.5 | 133.5 |
| ECI rank | #104 of 148 (best) | #106 of 148 | #119 of 148 |
| GPQA DiamondGraduate-level science questions | 71.7% (best) | 65.7% | 55.3% |
| OTIS Mock AIME 2024–2025Competition mathematics | 53.3% | 66.9% (best) | 8.5% |
| Price per million tokens | |||
| Input | $0.70 (best) | $0.70 (best) | $3.00 |
| Output | $2.60 (best) | $2.80 | $15.00 |
| Cached input | — | — | — |
| Blended (3:1) | $1.18 (best) | $1.23 | $6.00 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 11 providers | Official Alibaba API | Median of 1 providers |
| Limits | |||
| Context window | 128,000 tokens | 131,072 tokens | 200,000 tokens (best) |
| Max output | 32,768 tokens (best) | 16,384 tokens | 8,192 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | No | Yes |
| PDFs | No | No | Yes |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yes | Yes | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | Open | Open | Proprietary |
| API model ID | — | qwen3-32b | — |
| API providers | 12 | 14 (best) | 1 |
| Released | Jan 20, 2025 | Apr 29, 2025 | Oct 22, 2024 |
| Knowledge cutoff | Jul 2024 | Apr 2025 | Apr 30, 2024 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- DeepSeek-R1$12.20
- Qwen3 32B$12.60
- Claude Sonnet 3.5 v2$60.00
Which should you choose?
Which is better: DeepSeek-R1, Qwen3 32B or Claude Sonnet 3.5 v2?
It is close. Our weighted score puts them within a point (DeepSeek-R1 51/100, Qwen3 32B 51/100, Claude Sonnet 3.5 v2 44/100), so choose by what matters most for your work: DeepSeek-R1 for raw capability and Claude Sonnet 3.5 v2 for long inputs. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
Which is cheaper, DeepSeek-R1, Qwen3 32B or Claude Sonnet 3.5 v2?
DeepSeek-R1 is cheaper at $0.70 input / $2.60 output per million tokens (median across 11 API providers). Qwen3 32B costs $0.70 input / $2.80 output per million tokens (official Alibaba API price); Claude Sonnet 3.5 v2 costs $3.00 input / $15.00 output per million tokens (median across 1 API provider). At a typical mix of three input tokens to one output token, that is $1.18 per million tokens for DeepSeek-R1 versus $1.23 for Qwen3 32B (1× as much) and $6.00 for Claude Sonnet 3.5 v2 (5.1× as much).
Which scores higher on benchmarks?
DeepSeek-R1 scores higher on the Capabilities Index (ECI): DeepSeek-R1 139.0 (#104 of 148), Qwen3 32B 138.5 (#106 of 148) and Claude Sonnet 3.5 v2 133.5 (#119 of 148). The confidence ranges of the top two overlap (136.2–140.4 vs 135.1–140.4), so treat the gap as small. On individual benchmarks: GPQA Diamond — DeepSeek-R1 71.7%, Qwen3 32B 65.7%, Claude Sonnet 3.5 v2 55.3%; OTIS Mock AIME 2024–2025 — Qwen3 32B 66.9%, DeepSeek-R1 53.3%, Claude Sonnet 3.5 v2 8.5%.
Which is better for coding?
There are no published SWE-bench Verified results for DeepSeek-R1, Qwen3 32B and Claude Sonnet 3.5 v2 yet, so there is no like-for-like coding score. On overall capability, DeepSeek-R1 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?
Claude Sonnet 3.5 v2 has the largest context window at 200,000 tokens, against 131,072 for Qwen3 32B and 128,000 for DeepSeek-R1. Maximum output per response: DeepSeek-R1 up to 32,768, Qwen3 32B up to 16,384, Claude Sonnet 3.5 v2 up to 8,192 tokens.
Which can read images, PDFs, audio or video?
DeepSeek-R1 accepts text; Qwen3 32B accepts text; Claude Sonnet 3.5 v2 accepts text, images and PDFs. Claude Sonnet 3.5 v2 handles the widest range of inputs.
Are any of these open source?
DeepSeek-R1 and Qwen3 32B publishes its weights and can be self-hosted; Claude Sonnet 3.5 v2 is proprietary.
Which is newer?
Qwen3 32B is the newest, released Apr 29, 2025. DeepSeek-R1 came out Jan 20, 2025; Claude Sonnet 3.5 v2 came out Oct 22, 2024. Knowledge cutoff: DeepSeek-R1 Jul 2024, Qwen3 32B Apr 2025, Claude Sonnet 3.5 v2 Apr 30, 2024.
How do you decide the winner?
Each model gets a 0–100 score on 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 Oct 4, 2026.
