ZMIME
Comparison · 3 models · Updated Nov 2, 2026

Mistral 7B Instruct v0.3 vs Mixtral 8x22B Instruct vs Llama-3.1-8B-Instruct

Mixtral 8x22B Instruct comes out ahead, 37 to 32 and 22 on our weighted score, though Llama-3.1-8B-Instruct is 46× cheaper per token.

  1. Mistral AI

    Mistral 7B Instruct v0.3

    Released May 22, 2024

    22/100
    • ECI109.0
    • Price—
    • Context33K
    Full specs Remove Mistral 7B Instruct v0.3
  2. Our pick

    Mistral AI

    Mixtral 8x22B Instruct

    Released Apr 17, 2024

    37/100
    • ECI122.0
    • Price$2.00 / $6.00
    • Context66K
    Full specs Remove Mixtral 8x22B Instruct

  3. Meta

    Llama-3.1-8B-Instruct

    Released Jul 23, 2024

    32/100
    • ECI116.4
    • Price$0.06 / $0.08
    • Context128K
    Full specs Remove Llama-3.1-8B-Instruct
01 — Verdict

Mixtral 8x22B Instruct is our pick

Mixtral 8x22B Instruct is the better all-round choice, scoring 37/100 against Llama-3.1-8B-Instruct (32) and Mistral 7B Instruct v0.3 (22). It leads on capability and inputs & features. Llama-3.1-8B-Instruct wins on context window. The score weighs capability 67%, inputs & features 20%, context window 13%.

  • CapabilityMixtral 8x22B InstructCapabilities Index (ECI): Mixtral 8x22B Instruct 122.0 · Llama-3.1-8B-Instruct 116.4 · Mistral 7B Instruct v0.3 109.0
  • Lowest priceLlama-3.1-8B-InstructLlama-3.1-8B-Instruct $0.065 · Mixtral 8x22B Instruct $3.00 per 1M tokens (3:1 blend) · Mistral 7B Instruct v0.3 unpriced
  • Longest contextLlama-3.1-8B-InstructLlama-3.1-8B-Instruct 128,000 · Mixtral 8x22B Instruct 65,536 · Mistral 7B Instruct v0.3 32,768 tokens
  • Widest inputsMixtral 8x22B InstructMistral 7B Instruct v0.3: Text · Mixtral 8x22B Instruct: Text, PDFs · Llama-3.1-8B-Instruct: Text
  • Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
How the score is built
MeasureWeightMistral 7B Instruct v0.3Mixtral 8x22B InstructLlama-3.1-8B-Instruct
CapabilityCapabilities Index (ECI)67%264235
Inputs & features20%253525
Context window13%01224
Overall100%22/10037/10032/100

Left out because at least one model lacks the data: price. The remaining weights were rescaled.

02 — Side by side

Every spec in one table

Highlighted cells lead their row. Dashes mean the data is not published.

Mistral 7B Instruct v0.3 vs Mixtral 8x22B Instruct vs Llama-3.1-8B-Instruct specifications side by side
SpecificationMistral 7B Instruct v0.3Mistral AIMixtral 8x22B InstructMistral AILlama-3.1-8B-InstructMeta
Capability
Capabilities Index (ECI)109.0122.0 (best)116.4
ECI rank#164 of 165#155 of 165 (best)#162 of 165
GPQA DiamondGraduate-level science questions15.2%34.1% (best)27.0%
OTIS Mock AIME 2024–2025Competition mathematics0.3%—1.7% (best)
Price per million tokens
Input—$2.00$0.06 (best)
Output—$6.00$0.08 (best)
Cached input———
Blended (3:1)—$3.00$0.065 (best)
Long-context rate—Same rateSame rate
Price source—Median of 5 providersMedian of 17 providers
Limits
Context window32,768 tokens65,536 tokens128,000 tokens (best)
Max output32,768 tokens52,428 tokens (best)4,096 tokens
Inputs and features
TextYesYesYes
ImagesNoNoNo
PDFsNoYesNo
AudioNoNoNo
VideoNoNoNo
ReasoningNoNoNo
Tool callingYesYesYes
Structured outputNoNoNo
Availability
WeightsOpenApache 2.0OpenOpen
API model ID———
API providers1518 (best)
ReleasedMay 22, 2024Apr 17, 2024Jul 23, 2024
Knowledge cutoff—Jan 31, 2024Dec 2023
03 — Cost

What would a month cost?

Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.

  • Mistral 7B Instruct v0.3—
  • Mixtral 8x22B Instruct$32.00
  • Llama-3.1-8B-Instruct$0.76
04 — Questions

Which should you choose?

Which is better: Mistral 7B Instruct v0.3, Mixtral 8x22B Instruct or Llama-3.1-8B-Instruct?

Mixtral 8x22B Instruct is the better all-round choice, scoring 37/100 against Llama-3.1-8B-Instruct (32) and Mistral 7B Instruct v0.3 (22). It leads on capability and inputs & features. Llama-3.1-8B-Instruct wins on context window. The score weighs capability 67%, inputs & features 20%, context window 13%.

Which is cheaper, Mistral 7B Instruct v0.3, Mixtral 8x22B Instruct or Llama-3.1-8B-Instruct?

Llama-3.1-8B-Instruct is cheaper at $0.06 input / $0.08 output per million tokens (median across 17 API providers). Mixtral 8x22B Instruct costs $2.00 input / $6.00 output per million tokens (median across 5 API providers). At a typical mix of three input tokens to one output token, that is $0.065 per million tokens for Llama-3.1-8B-Instruct versus $3.00 for Mixtral 8x22B Instruct (46× as much). Mistral 7B Instruct v0.3 has no published per-token price.

Which scores higher on benchmarks?

Mixtral 8x22B Instruct scores higher on the Capabilities Index (ECI): Mixtral 8x22B Instruct 122.0 (#155 of 165), Llama-3.1-8B-Instruct 116.4 (#162 of 165) and Mistral 7B Instruct v0.3 109.0 (#164 of 165). The confidence ranges of the top two overlap (115.7–124.3 vs 107.0–121.6), so treat the gap as small. On individual benchmarks: GPQA Diamond — Mixtral 8x22B Instruct 34.1%, Llama-3.1-8B-Instruct 27.0%, Mistral 7B Instruct v0.3 15.2%.

Which is better for coding?

There are no published SWE-bench Verified results for Mistral 7B Instruct v0.3, Mixtral 8x22B Instruct and Llama-3.1-8B-Instruct yet, so there is no like-for-like coding score. On overall capability, Mixtral 8x22B 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?

Llama-3.1-8B-Instruct has the largest context window at 128,000 tokens, against 65,536 for Mixtral 8x22B Instruct and 32,768 for Mistral 7B Instruct v0.3. Maximum output per response: Mistral 7B Instruct v0.3 up to 32,768, Mixtral 8x22B Instruct up to 52,428, Llama-3.1-8B-Instruct up to 4,096 tokens.

Which can read images, PDFs, audio or video?

Mistral 7B Instruct v0.3 accepts text; Mixtral 8x22B Instruct accepts text and PDFs; Llama-3.1-8B-Instruct accepts text. Mixtral 8x22B Instruct handles the widest range of inputs.

Are any of these open source?

Yes, all three publish their weights (Apache 2.0), so you can self-host them.

Which is newer?

Llama-3.1-8B-Instruct is the newest, released Jul 23, 2024. Mistral 7B Instruct v0.3 came out May 22, 2024; Mixtral 8x22B Instruct came out Apr 17, 2024. Knowledge cutoff: Mixtral 8x22B Instruct Jan 31, 2024, Llama-3.1-8B-Instruct Dec 2023.

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 Nov 2, 2026.

05 — Keep comparing

Related comparisons