Claude Haiku 3.5 vs Nova Pro vs Qwen2.5 32B Instruct
Too close to call on our weighted score (Nova Pro 49, Claude Haiku 3.5 49, Qwen2.5 32B Instruct 42). The right pick depends on what you value most.
AnthropicClaude Haiku 3.5
49/100- ECI127.2
- Price—
- Context200K
AmazonNova Pro
49/100- ECI123.8
- Price$0.80 / $3.20
- Context300K
Alibaba (Qwen)Qwen2.5 32B Instruct
42/100- ECI128.5
- Price$0.70 / $2.80
- Context131K
Too close to call
It is close. Our weighted score puts them within a point (Nova Pro 49/100, Claude Haiku 3.5 49/100, Qwen2.5 32B Instruct 42/100), so choose by what matters most for your work: Qwen2.5 32B Instruct for raw capability and Nova Pro for long inputs. The score weighs capability 67%, inputs & features 20%, context window 13%.
- CapabilityQwen2.5 32B InstructCapabilities Index (ECI): Qwen2.5 32B Instruct 128.5 · Claude Haiku 3.5 127.2 · Nova Pro 123.8
- Lowest priceQwen2.5 32B InstructQwen2.5 32B Instruct $1.23 · Nova Pro $1.40 per 1M tokens (3:1 blend) · Claude Haiku 3.5 unpriced
- Longest contextNova ProNova Pro 300,000 · Claude Haiku 3.5 200,000 · Qwen2.5 32B Instruct 131,072 tokens
- Widest inputsNova ProClaude Haiku 3.5: Text, Images, PDFs · Nova Pro: Text, Images, PDFs, Video · Qwen2.5 32B Instruct: Text
- Self-hostingQwen2.5 32B InstructPublishes downloadable weights
| Measure | Weight | Claude Haiku 3.5 | Nova Pro | Qwen2.5 32B Instruct |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 67% | 49 | 45 | 51 |
| Inputs & features | 20% | 60 | 70 | 25 |
| Context window | 13% | 32 | 39 | 24 |
| Overall | 100% | 49/100 | 49/100 | 42/100 |
Left out because at least one model lacks the data: price. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | Claude Haiku 3.5Anthropic | Nova ProAmazon | Qwen2.5 32B InstructAlibaba (Qwen) |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 127.2 | 123.8 | 128.5 (best) |
| ECI rank | #134 of 148 | #137 of 148 | #131 of 148 (best) |
| GPQA DiamondGraduate-level science questions | 38.1% | — | 46.1% (best) |
| OTIS Mock AIME 2024–2025Competition mathematics | 4.3% | — | 7.4% (best) |
| Price per million tokens | |||
| Input | — | $0.80 | $0.70 (best) |
| Output | — | $3.20 | $2.80 (best) |
| Cached input | — | $0.20 | — |
| Blended (3:1) | — | $1.40 | $1.23 (best) |
| Long-context rate | — | Same rate | Same rate |
| Price source | — | Official Amazon Bedrock API | Official Alibaba API |
| Limits | |||
| Context window | 200,000 tokens | 300,000 tokens (best) | 131,072 tokens |
| Max output | 8,192 tokens | 10,000 tokens (best) | 8,192 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | Yes | No |
| PDFs | Yes | Yes | No |
| Audio | No | No | No |
| Video | No | Yes | No |
| Reasoning | No | No | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | Proprietary | Proprietary | Open |
| API model ID | — | amazon.nova-pro-v1:0 | qwen2-5-32b-instruct |
| API providers | — | 3 (best) | 1 |
| Released | Oct 22, 2024 | Dec 3, 2024 | Sep 17, 2024 |
| Knowledge cutoff | Jul 31, 2024 | Oct 2024 | Apr 2024 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Claude Haiku 3.5—
- Nova Pro$14.40
- Qwen2.5 32B Instruct$12.60
Which should you choose?
Which is better: Claude Haiku 3.5, Nova Pro or Qwen2.5 32B Instruct?
It is close. Our weighted score puts them within a point (Nova Pro 49/100, Claude Haiku 3.5 49/100, Qwen2.5 32B Instruct 42/100), so choose by what matters most for your work: Qwen2.5 32B Instruct for raw capability and Nova Pro for long inputs. The score weighs capability 67%, inputs & features 20%, context window 13%.
Which is cheaper, Claude Haiku 3.5, Nova Pro or Qwen2.5 32B Instruct?
Qwen2.5 32B Instruct is cheaper at $0.70 input / $2.80 output per million tokens (official Alibaba API price). Nova Pro costs $0.80 input / $3.20 output per million tokens (official Amazon Bedrock API price). At a typical mix of three input tokens to one output token, that is $1.23 per million tokens for Qwen2.5 32B Instruct versus $1.40 for Nova Pro (1.1× as much). Claude Haiku 3.5 has no published per-token price.
Which scores higher on benchmarks?
Qwen2.5 32B Instruct scores higher on the Capabilities Index (ECI): Qwen2.5 32B Instruct 128.5 (#131 of 148), Claude Haiku 3.5 127.2 (#134 of 148) and Nova Pro 123.8 (#137 of 148). The confidence ranges of the top two overlap (123.5–130.0 vs 120.7–129.7), so treat the gap as small.
Which is better for coding?
There are no published SWE-bench Verified results for Claude Haiku 3.5, Nova Pro and Qwen2.5 32B Instruct yet, so there is no like-for-like coding score. On overall capability, Qwen2.5 32B Instruct 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?
Nova Pro has the largest context window at 300,000 tokens, against 200,000 for Claude Haiku 3.5 and 131,072 for Qwen2.5 32B Instruct. Maximum output per response: Claude Haiku 3.5 up to 8,192, Nova Pro up to 10,000, Qwen2.5 32B Instruct up to 8,192 tokens.
Which can read images, PDFs, audio or video?
Claude Haiku 3.5 accepts text, images and PDFs; Nova Pro accepts text, images, PDFs and video; Qwen2.5 32B Instruct accepts text. Nova Pro handles the widest range of inputs.
Are any of these open source?
Qwen2.5 32B Instruct publishes its weights and can be self-hosted; Claude Haiku 3.5 and Nova Pro is proprietary.
Which is newer?
Nova Pro is the newest, released Dec 3, 2024. Claude Haiku 3.5 came out Oct 22, 2024; Qwen2.5 32B Instruct came out Sep 17, 2024. Knowledge cutoff: Claude Haiku 3.5 Jul 31, 2024, Nova Pro Oct 2024, Qwen2.5 32B Instruct Apr 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.
Related comparisons
- Claude Haiku 3.5 vs Llama-3.3-70B-InstructMeta · ECI 127.3
- Claude Haiku 3.5 vs Mistral Large 2.1Mistral AI · ECI 128.5
- Claude Haiku 3.5 vs GPT-4o miniOpenAI · ECI 126.6
- Nova Pro vs Llama-3.3-70B-InstructMeta · ECI 127.3
- Nova Pro vs Mistral Large 2.1Mistral AI · ECI 128.5
- Qwen2.5 32B Instruct vs Mistral Large 2.1Mistral AI · ECI 128.5
