Llama-3.3-70B-Instruct vs Nova Pro vs Qwen2.5 32B Instruct
Too close to call on our weighted score (Nova Pro 48, Llama-3.3-70B-Instruct 46, Qwen2.5 32B Instruct 43). The right pick depends on what you value most.
MetaLlama-3.3-70B-Instruct
46/100- ECI127.3
- Price$0.59 / $0.724
- Context128K
AmazonNova Pro
48/100- ECI123.8
- Price$0.80 / $3.20
- Context300K
Alibaba (Qwen)Qwen2.5 32B Instruct
43/100- ECI128.5
- Price$0.70 / $2.80
- Context131K
Too close to call
It is close. Our weighted score puts them within 2 points (Nova Pro 48/100, Llama-3.3-70B-Instruct 46/100, Qwen2.5 32B Instruct 43/100), so choose by what matters most for your work: Qwen2.5 32B Instruct for raw capability, Llama-3.3-70B-Instruct on price and Nova Pro for long inputs. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
- CapabilityQwen2.5 32B InstructCapabilities Index (ECI): Qwen2.5 32B Instruct 128.5 · Llama-3.3-70B-Instruct 127.3 · Nova Pro 123.8
- Lowest priceLlama-3.3-70B-InstructLlama-3.3-70B-Instruct $0.624 · Qwen2.5 32B Instruct $1.23 · Nova Pro $1.40 per 1M tokens (3:1 blend)
- Longest contextNova ProNova Pro 300,000 · Qwen2.5 32B Instruct 131,072 · Llama-3.3-70B-Instruct 128,000 tokens
- Widest inputsNova ProLlama-3.3-70B-Instruct: Text · Nova Pro: Text, Images, PDFs, Video · Qwen2.5 32B Instruct: Text
- Self-hostingLlama-3.3-70B-Instruct and Qwen2.5 32B InstructPublishes downloadable weights
| Measure | Weight | Llama-3.3-70B-Instruct | Nova Pro | Qwen2.5 32B Instruct |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 50% | 49 | 45 | 51 |
| Price | 25% | 60 | 43 | 46 |
| Inputs & features | 15% | 25 | 70 | 25 |
| Context window | 10% | 24 | 39 | 24 |
| Overall | 100% | 46/100 | 48/100 | 43/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | Llama-3.3-70B-InstructMeta | Nova ProAmazon | Qwen2.5 32B InstructAlibaba (Qwen) |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 127.3 | 123.8 | 128.5 (best) |
| ECI rank | #133 of 148 | #137 of 148 | #131 of 148 (best) |
| GPQA DiamondGraduate-level science questions | 47.4% (best) | — | 46.1% |
| OTIS Mock AIME 2024–2025Competition mathematics | 5.1% | — | 7.4% (best) |
| Price per million tokens | |||
| Input | $0.59 (best) | $0.80 | $0.70 |
| Output | $0.724 (best) | $3.20 | $2.80 |
| Cached input | — | $0.20 | — |
| Blended (3:1) | $0.624 (best) | $1.40 | $1.23 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 21 providers | Official Amazon Bedrock API | Official Alibaba API |
| Limits | |||
| Context window | 128,000 tokens | 300,000 tokens (best) | 131,072 tokens |
| Max output | 4,096 tokens | 10,000 tokens (best) | 8,192 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | Yes | No |
| PDFs | No | 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 | Open | Proprietary | Open |
| API model ID | llama-3.3-70b-instruct | amazon.nova-pro-v1:0 | qwen2-5-32b-instruct |
| API providers | 24 (best) | 3 | 1 |
| Released | Dec 6, 2024 | Dec 3, 2024 | Sep 17, 2024 |
| Knowledge cutoff | Dec 2023 | 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.
- Llama-3.3-70B-Instruct$7.35
- Nova Pro$14.40
- Qwen2.5 32B Instruct$12.60
Which should you choose?
Which is better: Llama-3.3-70B-Instruct, Nova Pro or Qwen2.5 32B Instruct?
It is close. Our weighted score puts them within 2 points (Nova Pro 48/100, Llama-3.3-70B-Instruct 46/100, Qwen2.5 32B Instruct 43/100), so choose by what matters most for your work: Qwen2.5 32B Instruct for raw capability, Llama-3.3-70B-Instruct on price and Nova Pro for long inputs. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
Which is cheaper, Llama-3.3-70B-Instruct, Nova Pro or Qwen2.5 32B Instruct?
Llama-3.3-70B-Instruct is cheaper at $0.59 input / $0.724 output per million tokens (median across 21 API providers; free on Llama). Qwen2.5 32B Instruct costs $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 $0.624 per million tokens for Llama-3.3-70B-Instruct versus $1.23 for Qwen2.5 32B Instruct (2× as much) and $1.40 for Nova Pro (2.2× as much).
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), Llama-3.3-70B-Instruct 127.3 (#133 of 148) and Nova Pro 123.8 (#137 of 148). The confidence ranges of the top two overlap (123.5–130.0 vs 122.5–129.5), so treat the gap as small.
Which is better for coding?
There are no published SWE-bench Verified results for Llama-3.3-70B-Instruct, 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 131,072 for Qwen2.5 32B Instruct and 128,000 for Llama-3.3-70B-Instruct. Maximum output per response: Llama-3.3-70B-Instruct up to 4,096, Nova Pro up to 10,000, Qwen2.5 32B Instruct up to 8,192 tokens.
Which can read images, PDFs, audio or video?
Llama-3.3-70B-Instruct accepts text; 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?
Llama-3.3-70B-Instruct and Qwen2.5 32B Instruct publishes its weights and can be self-hosted; Nova Pro is proprietary.
Which is newer?
Llama-3.3-70B-Instruct is the newest, released Dec 6, 2024. Nova Pro came out Dec 3, 2024; Qwen2.5 32B Instruct came out Sep 17, 2024. Knowledge cutoff: Llama-3.3-70B-Instruct Dec 2023, 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
- Llama-3.3-70B-Instruct vs Mistral Small 3.1 24BMistral AI · ECI 127.5
- Llama-3.3-70B-Instruct vs DeepSeek-V3DeepSeek · ECI 132.3
- Llama-3.3-70B-Instruct vs Claude Haiku 3.5Anthropic · ECI 127.2
- Nova Pro vs Mistral Large 2.1Mistral AI · ECI 128.5
- Nova Pro vs Claude Haiku 3.5Anthropic · ECI 127.2
- Qwen2.5 32B Instruct vs Mistral Large 2.1Mistral AI · ECI 128.5
