Mistral Small 3.1 24B vs Hy3 preview vs Llama 4 Scout 17B Instruct
Llama 4 Scout 17B Instruct comes out ahead, 71 to 61 and 56 on our weighted score, though Hy3 preview is 18% cheaper per token.
Mistral AIMistral Small 3.1 24B
61/100- ECI127.5
- Price$0.229 / $0.436
- Context128K
TencentHy3 preview
56/100- ECI—
- Price$0.176 / $0.586
- Context256K
- Our pick
MetaLlama 4 Scout 17B Instruct
71/100- ECI129.7
- Price$0.225 / $0.69
- Context10M
Llama 4 Scout 17B Instruct is our pick
Llama 4 Scout 17B Instruct is the better all-round choice, scoring 71/100 against Mistral Small 3.1 24B (61) and Hy3 preview (56). It leads on context window. Mistral Small 3.1 24B wins on inputs & features. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceHy3 previewHy3 preview $0.279 · Mistral Small 3.1 24B $0.281 · Llama 4 Scout 17B Instruct $0.341 per 1M tokens (3:1 blend)
- Longest contextLlama 4 Scout 17B InstructLlama 4 Scout 17B Instruct 10,000,000 · Hy3 preview 256,000 · Mistral Small 3.1 24B 128,000 tokens
- Widest inputsMistral Small 3.1 24B and Llama 4 Scout 17B InstructMistral Small 3.1 24B: Text, Images · Hy3 preview: Text · Llama 4 Scout 17B Instruct: Text, Images
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | Mistral Small 3.1 24B | Hy3 preview | Llama 4 Scout 17B Instruct |
|---|---|---|---|---|
| Price | 50% | 76 | 76 | 72 |
| Inputs & features | 30% | 60 | 35 | 50 |
| Context window | 20% | 24 | 36 | 100 |
| Overall | 100% | 61/100 | 56/100 | 71/100 |
Left out because at least one model lacks the data: capability. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | Mistral Small 3.1 24BMistral AI | Hy3 previewTencent | Llama 4 Scout 17B InstructMeta |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 127.5 | — | 129.7 (best) |
| ECI rank | #132 of 148 | — | #126 of 148 (best) |
| GPQA DiamondGraduate-level science questions | 47.5% | — | 51.8% (best) |
| OTIS Mock AIME 2024–2025Competition mathematics | 5.8% | — | 7.8% (best) |
| Price per million tokens | |||
| Input | $0.229 | $0.176 (best) | $0.225 |
| Output | $0.436 (best) | $0.586 | $0.69 |
| Cached input | — | — | — |
| Blended (3:1) | $0.281 | $0.279 (best) | $0.341 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 2 providers | Median of 4 providers | Median of 4 providers |
| Limits | |||
| Context window | 128,000 tokens | 256,000 tokens | 10,000,000 tokens (best) |
| Max output | 16,384 tokens | 64,000 tokens (best) | 16,384 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | No | Yes |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | No | Yeslow · medium · high | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | No | No |
| Availability | |||
| Weights | Open | Open | Open |
| API model ID | — | hy3-preview | — |
| API providers | 2 | 5 (best) | 4 |
| Released | Mar 17, 2025 | Apr 20, 2026 | Apr 5, 2025 |
| Knowledge cutoff | Jun 2024 | — | Aug 2024 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Mistral Small 3.1 24B$3.16
- Hy3 preview$2.93
- Llama 4 Scout 17B Instruct$3.63
Which should you choose?
Which is better: Mistral Small 3.1 24B, Hy3 preview or Llama 4 Scout 17B Instruct?
Llama 4 Scout 17B Instruct is the better all-round choice, scoring 71/100 against Mistral Small 3.1 24B (61) and Hy3 preview (56). It leads on context window. Mistral Small 3.1 24B wins on inputs & features. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Mistral Small 3.1 24B, Hy3 preview or Llama 4 Scout 17B Instruct?
Hy3 preview is cheaper at $0.176 input / $0.586 output per million tokens (median across 4 API providers; free on Tencent TokenHub). Mistral Small 3.1 24B costs $0.229 input / $0.436 output per million tokens (median across 2 API providers); Llama 4 Scout 17B Instruct costs $0.225 input / $0.69 output per million tokens (median across 4 API providers). At a typical mix of three input tokens to one output token, that is $0.279 per million tokens for Hy3 preview versus $0.281 for Mistral Small 3.1 24B (1× as much) and $0.341 for Llama 4 Scout 17B Instruct (1.2× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Mistral Small 3.1 24B has an ECI of 127.5, Hy3 preview has not been scored yet and Llama 4 Scout 17B Instruct has an ECI of 129.7.
Which is better for coding?
There are no published SWE-bench Verified results for Mistral Small 3.1 24B, Hy3 preview and Llama 4 Scout 17B Instruct yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.
Which has the bigger context window?
Llama 4 Scout 17B Instruct has the largest context window at 10,000,000 tokens, against 256,000 for Hy3 preview and 128,000 for Mistral Small 3.1 24B. Maximum output per response: Mistral Small 3.1 24B up to 16,384, Hy3 preview up to 64,000, Llama 4 Scout 17B Instruct up to 16,384 tokens.
Which can read images, PDFs, audio or video?
Mistral Small 3.1 24B accepts text and images; Hy3 preview accepts text; Llama 4 Scout 17B Instruct accepts text and images. Mistral Small 3.1 24B handles the widest range of inputs.
Are any of these open source?
Yes, all three publish their weights, so you can self-host them.
Which is newer?
Hy3 preview is the newest, released Apr 20, 2026. Llama 4 Scout 17B Instruct came out Apr 5, 2025; Mistral Small 3.1 24B came out Mar 17, 2025. Knowledge cutoff: Mistral Small 3.1 24B Jun 2024, Llama 4 Scout 17B Instruct Aug 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.
