Ministral 8B Instruct vs Mistral Large 2.1 vs Llama-3.1-8B-Instruct
Mistral Large 2.1 comes out ahead, 48 to 28 and 28 on our weighted score, though Llama-3.1-8B-Instruct is 49× cheaper per token.
Mistral AI
Ministral 8B Instruct
28/100- ECI—
- Price—
- Context131K
- Our pick
Mistral AI
Mistral Large 2.1
48/100- ECI—
- Price$2.00 / $6.00
- Context131K
Meta
Llama-3.1-8B-Instruct
28/100- ECI110.9
- Price$0.055 / $0.078
- Context128K
Mistral AI
Ministral 8B Instruct
28/100- ECI—
- Price—
- Context131K
- Our pick
Mistral AI
Mistral Large 2.1
48/100- ECI—
- Price$2.00 / $6.00
- Context131K
Meta
Llama-3.1-8B-Instruct
28/100- ECI110.9
- Price$0.055 / $0.078
- Context128K
Mistral Large 2.1 is our pick
Mistral Large 2.1 is the better all-round choice, scoring 48/100 against Ministral 8B Instruct (28) and Llama-3.1-8B-Instruct (28). It leads on capability. The score weighs capability 83%, context window 17%. Capability uses the average of shared benchmarks (GPQA Diamond), because Ministral 8B Instruct and Mistral Large 2.1 has no Capabilities Index score yet.
- CapabilityMistral Large 2.1Shared benchmarks: Mistral Large 2.1 51.3% · Ministral 8B Instruct 27.2% · Llama-3.1-8B-Instruct 27.0%
- Lowest priceLlama-3.1-8B-InstructLlama-3.1-8B-Instruct $0.061 · Mistral Large 2.1 $3.00 per 1M tokens (3:1 blend) · Ministral 8B Instruct unpriced
- Longest contextMinistral 8B Instruct and Mistral Large 2.1Ministral 8B Instruct 131,072 · Mistral Large 2.1 131,072 · Llama-3.1-8B-Instruct 128,000 tokens
- Widest inputsSame inputsMinistral 8B Instruct: Text · Mistral Large 2.1: Text · Llama-3.1-8B-Instruct: Text
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | Ministral 8B Instruct | Mistral Large 2.1 | Llama-3.1-8B-Instruct |
|---|---|---|---|---|
| CapabilityShared benchmarks | 83% | 27 | 51 | 27 |
| Context window | 17% | 32 | 32 | 32 |
| Overall | 100% | 28/100 | 48/100 | 28/100 |
Left out because at least one model lacks the data: price and inputs & features. The remaining weights were rescaled.
| Measure | Weight | Ministral 8B Instruct | Mistral Large 2.1 | Llama-3.1-8B-Instruct |
|---|---|---|---|---|
| CapabilityShared benchmarks | 83% | 27 | 51 | 27 |
| Context window | 17% | 32 | 32 | 32 |
| Overall | 100% | 28/100 | 48/100 | 28/100 |
Left out because at least one model lacks the data: inputs & features. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Unavailable specifications are marked “Not published”. Publisher results include their evaluation settings and source.
| Specification | |||
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | Not published | Not published | 110.9 |
| ECI rank | Not published | Not published | #158 of 159 |
| ECI confidence interval | Not published | Not published | 105.1–115.9 |
| ECI evaluation date | Not published | Not published | Jul 23, 2024 |
| Documented capabilities | Efficient open Mistral edge model for on-device chat and function calling | Flagship Mistral model for advanced reasoning, coding, and multilingual work | Compact open Llama model for lightweight chat, drafting, and self-hosting |
| Model parameters | Not published | Not published | Not published |
| Output modalities | Text | Text | Text |
| GPQA DiamondGraduate-level science questions | 27.2% | 51.3% (best) | 27.0% |
| OTIS Mock AIME 2024–2025Competition mathematics | Not published | 7.8% (best) | 1.7% |
| Publisher-reported benchmarks | |||
| Artificial Analysis Coding IndexPublisher-reported result | Not published | 13.8indexReported resultMar 11, 2026 | Not published |
| SciCodePublisher-reported result | Not published | 29.2percent correctReported resultMar 11, 2026 | Not published |
| Terminal-Bench HardPublisher-reported result | Not published | 6.1success rateReported resultMar 11, 2026 | Not published |
| Price per million tokens | |||
| Input | Not published | $2.00 | $0.055 (best) |
| Output | Not published | $6.00 | $0.078 (best) |
| Cached input | Not published | Not published | Not published |
| Blended (3:1) | Not published | $3.00 | $0.061 (best) |
| Long-context rate | Not published | Same rate | Same rate |
| Price source | Not published | Median of 2 providers | Median of 18 providers |
| Limits | |||
| Context window | 131,072 tokens (best) | 131,072 tokens (best) | 128,000 tokens |
| Max input | Not published | Not published | Not published |
| Max output | 8,192 tokens | 16,384 tokens (best) | 4,096 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | No | No |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | No | No | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | Not published | Not published | Not published |
| File attachments | No | No | No |
| Temperature control | Yes | Yes | Yes |
| Reasoning control | Not published | Not published | Not published |
| Availability | |||
| Weights | OpenMistral Research License | Open | Open |
| API model ID | Not published | Not published | Not published |
| API providers | Not published | 2 | 19 (best) |
| Released | Oct 16, 2024 | Nov 18, 2024 | Jul 23, 2024 |
| Knowledge cutoff | Not published | Nov 2024 | Dec 2023 |
| Publisher documentation | Model weights | Model weights | Model weights |
| Source metadata updated | Oct 16, 2024 | Nov 18, 2024 | Jul 23, 2024 |
| Data sources | models.dev · Epoch AI | models.dev · Epoch AI | models.dev · Epoch AI |
| Catalog checked | Nov 22, 2026 | Nov 22, 2026 | Nov 22, 2026 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
Ministral 8B Instruct—
Mistral Large 2.1$32.00
Llama-3.1-8B-Instruct$0.707
Which should you choose?
Which is better: Ministral 8B Instruct, Mistral Large 2.1 or Llama-3.1-8B-Instruct?
Mistral Large 2.1 is the better all-round choice, scoring 48/100 against Ministral 8B Instruct (28) and Llama-3.1-8B-Instruct (28). It leads on capability. The score weighs capability 83%, context window 17%. Capability uses the average of shared benchmarks (GPQA Diamond), because Ministral 8B Instruct and Mistral Large 2.1 has no Capabilities Index score yet.
Which is cheaper, Ministral 8B Instruct, Mistral Large 2.1 or Llama-3.1-8B-Instruct?
Llama-3.1-8B-Instruct is cheaper at $0.055 input / $0.078 output per million tokens (median across 18 API providers). Mistral Large 2.1 costs $2.00 input / $6.00 output per million tokens (median across 2 API providers). At a typical mix of three input tokens to one output token, that is $0.061 per million tokens for Llama-3.1-8B-Instruct versus $3.00 for Mistral Large 2.1 (49× as much). Ministral 8B Instruct has no published per-token price.
Which scores higher on benchmarks?
Not every model here has a Capabilities Index score, so we compare the average of shared benchmarks (GPQA Diamond): Mistral Large 2.1 51.3%, Ministral 8B Instruct 27.2% and Llama-3.1-8B-Instruct 27.0%. On individual benchmarks: GPQA Diamond — Mistral Large 2.1 51.3%, Ministral 8B Instruct 27.2%, Llama-3.1-8B-Instruct 27.0%.
Which is better for coding?
There are no published SWE-bench Verified results for Ministral 8B Instruct, Mistral Large 2.1 and Llama-3.1-8B-Instruct yet, so there is no like-for-like coding score. On overall capability, Mistral Large 2.1 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?
Ministral 8B Instruct and Mistral Large 2.1 have the largest context windows (131,072 and 131,072 tokens), against 128,000 for Llama-3.1-8B-Instruct. Maximum output per response: Ministral 8B Instruct up to 8,192, Mistral Large 2.1 up to 16,384, Llama-3.1-8B-Instruct up to 4,096 tokens.
Which can read images, PDFs, audio or video?
Ministral 8B Instruct accepts text; Mistral Large 2.1 accepts text; Llama-3.1-8B-Instruct accepts text. They handle the same number of input types.
Are any of these open source?
Yes, all three publish their weights (Mistral Research License), so you can self-host them.
Which is newer?
Mistral Large 2.1 is the newest, released Nov 18, 2024. Ministral 8B Instruct came out Oct 16, 2024; Llama-3.1-8B-Instruct came out Jul 23, 2024. Knowledge cutoff: Mistral Large 2.1 Nov 2024, Llama-3.1-8B-Instruct Dec 2023.
How do you decide the winner?
This rating uses capability (83%, independent benchmark results); context window (17%, 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 22, 2026.