QwQ 32B vs Llama-3.1-70B-Instruct vs DeepSeek V3 0324
Too close to call on our weighted score (QwQ 32B 55, DeepSeek V3 0324 55, Llama-3.1-70B-Instruct 47). The right pick depends on what you value most.
Price in rating: Included · Exclude price
Alibaba (Qwen)
QwQ 32B
55/100- ECI136.5
- Price$0.53 / $0.70
- Context131K
Meta
Llama-3.1-70B-Instruct
47/100- ECI124.7
- Price$0.72 / $0.72
- Context128K
DeepSeek
DeepSeek V3 0324
55/100- ECI135.3
- Price$0.275 / $1.12
- Context164K
Too close to call
It is close. Our weighted score puts them within a point (QwQ 32B 55/100, DeepSeek V3 0324 55/100, Llama-3.1-70B-Instruct 47/100), so choose by what matters most for your work: QwQ 32B for raw capability, DeepSeek V3 0324 on price and DeepSeek V3 0324 for long inputs. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
- CapabilityQwQ 32BCapabilities Index (ECI): QwQ 32B 136.5 · DeepSeek V3 0324 135.3 · Llama-3.1-70B-Instruct 124.7
- Lowest priceDeepSeek V3 0324DeepSeek V3 0324 $0.486 · QwQ 32B $0.573 · Llama-3.1-70B-Instruct $0.72 per 1M tokens (3:1 blend)
- Longest contextDeepSeek V3 0324DeepSeek V3 0324 163,840 · QwQ 32B 131,072 · Llama-3.1-70B-Instruct 128,000 tokens
- Widest inputsSame inputsQwQ 32B: Text · Llama-3.1-70B-Instruct: Text · DeepSeek V3 0324: Text
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | QwQ 32B | Llama-3.1-70B-Instruct | DeepSeek V3 0324 |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 50% | 65 | 53 | 63 |
| Price | 25% | 59 | 54 | 62 |
| Inputs & features | 15% | 35 | 25 | 25 |
| Context window | 10% | 32 | 32 | 36 |
| Overall | 100% | 55/100 | 47/100 | 55/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | |||
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 136.5 (best) | 124.7 | 135.3 |
| ECI rank | #122 of 158 (best) | #146 of 158 | #125 of 158 |
| GPQA DiamondGraduate-level science questions | 65.3% | 44.2% | 67.6% (best) |
| OTIS Mock AIME 2024–2025Competition mathematics | 59.2% (best) | 3.6% | 37.8% |
| Price per million tokens | |||
| Input | $0.53 | $0.72 | $0.275 (best) |
| Output | $0.70 (best) | $0.72 | $1.12 |
| Cached input | — | — | — |
| Blended (3:1) | $0.573 | $0.72 | $0.486 (best) |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 2 providers | Median of 7 providers | Median of 13 providers |
| Limits | |||
| Context window | 131,072 tokens | 128,000 tokens | 163,840 tokens (best) |
| Max output | 8,192 tokens | 4,096 tokens | 163,840 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | No | No |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yes | No | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | Open | Open | Open |
| API model ID | — | — | — |
| API providers | 2 | 7 | 13 (best) |
| Released | Mar 5, 2025 | Jul 23, 2024 | Mar 24, 2025 |
| Knowledge cutoff | Apr 2024 | Dec 2023 | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
QwQ 32B$6.70
Llama-3.1-70B-Instruct$8.64
DeepSeek V3 0324$4.99
Which should you choose?
Which is better: QwQ 32B, Llama-3.1-70B-Instruct or DeepSeek V3 0324?
It is close. Our weighted score puts them within a point (QwQ 32B 55/100, DeepSeek V3 0324 55/100, Llama-3.1-70B-Instruct 47/100), so choose by what matters most for your work: QwQ 32B for raw capability, DeepSeek V3 0324 on price and DeepSeek V3 0324 for long inputs. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
Which is cheaper, QwQ 32B, Llama-3.1-70B-Instruct or DeepSeek V3 0324?
DeepSeek V3 0324 is cheaper at $0.275 input / $1.12 output per million tokens (median across 13 API providers). QwQ 32B costs $0.53 input / $0.70 output per million tokens (median across 2 API providers); Llama-3.1-70B-Instruct costs $0.72 input / $0.72 output per million tokens (median across 7 API providers). At a typical mix of three input tokens to one output token, that is $0.486 per million tokens for DeepSeek V3 0324 versus $0.573 for QwQ 32B (1.2× as much) and $0.72 for Llama-3.1-70B-Instruct (1.5× as much).
Which scores higher on benchmarks?
QwQ 32B scores higher on the Capabilities Index (ECI): QwQ 32B 136.5 (#122 of 158), DeepSeek V3 0324 135.3 (#125 of 158) and Llama-3.1-70B-Instruct 124.7 (#146 of 158). The confidence ranges of the top two overlap (133.7–139.2 vs 133.1–137.3), so treat the gap as small. On individual benchmarks: GPQA Diamond — DeepSeek V3 0324 67.6%, QwQ 32B 65.3%, Llama-3.1-70B-Instruct 44.2%; OTIS Mock AIME 2024–2025 — QwQ 32B 59.2%, DeepSeek V3 0324 37.8%, Llama-3.1-70B-Instruct 3.6%.
Which is better for coding?
There are no published SWE-bench Verified results for QwQ 32B, Llama-3.1-70B-Instruct and DeepSeek V3 0324 yet, so there is no like-for-like coding score. On overall capability, QwQ 32B 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?
DeepSeek V3 0324 has the largest context window at 163,840 tokens, against 131,072 for QwQ 32B and 128,000 for Llama-3.1-70B-Instruct. Maximum output per response: QwQ 32B up to 8,192, Llama-3.1-70B-Instruct up to 4,096, DeepSeek V3 0324 up to 163,840 tokens.
Which can read images, PDFs, audio or video?
QwQ 32B accepts text; Llama-3.1-70B-Instruct accepts text; DeepSeek V3 0324 accepts text. They handle the same number of input types.
Are any of these open source?
Yes, all three publish their weights, so you can self-host them.
Which is newer?
DeepSeek V3 0324 is the newest, released Mar 24, 2025. QwQ 32B came out Mar 5, 2025; Llama-3.1-70B-Instruct came out Jul 23, 2024. Knowledge cutoff: QwQ 32B Apr 2024, Llama-3.1-70B-Instruct Dec 2023.
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.