ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Mixtral 8x22B vs WizardLM 2 8x22B vs GPT-4o

GPT-4o comes out ahead, 43 to 39 and 29 on our weighted score, though WizardLM 2 8x22B is 8.9× cheaper per token.

Price in rating: Included · Exclude price

Retired Mixtral 8x22B is no longer offered by any provider we track. Its figures are the last published listing, kept so this comparison stays available.

  1. Mistral AI

    Mixtral 8x22B

    Released Apr 17, 2024Retired

    29/100
    • ECI122.0
    • Price$2.00 / $6.00
    • Context64K
  2. Microsoft

    WizardLM 2 8x22B

    Released Apr 15, 2024

    39/100
    • ECI—
    • Price$0.493 / $0.493
    • Context66K
  3. Our pick

    OpenAI

    GPT-4o

    Released May 13, 2024

    43/100
    • ECI128.1
    • Price$2.50 / $10.00
    • Context128K
01 — Verdict

GPT-4o is our pick

GPT-4o is the better all-round choice, scoring 43/100 against WizardLM 2 8x22B (39) and Mixtral 8x22B (29). It leads on capability, inputs & features and context window. WizardLM 2 8x22B wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%. Capability uses the average of shared benchmarks (GPQA Diamond), because WizardLM 2 8x22B has no Capabilities Index score yet.

  • CapabilityGPT-4oShared benchmarks: GPT-4o 48.9% · WizardLM 2 8x22B 43.4% · Mixtral 8x22B 34.1%
  • Lowest priceWizardLM 2 8x22BWizardLM 2 8x22B $0.493 · Mixtral 8x22B $3.00 · GPT-4o $4.38 per 1M tokens (3:1 blend)
  • Longest contextGPT-4oGPT-4o 128,000 · WizardLM 2 8x22B 65,536 · Mixtral 8x22B 64,000 tokens
  • Widest inputsGPT-4oMixtral 8x22B: Text · WizardLM 2 8x22B: Text · GPT-4o: Text, Images, PDFs
  • Self-hostingMixtral 8x22B and WizardLM 2 8x22BPublishes downloadable weights (Apache-2.0)
How the score is built
MeasureWeightMixtral 8x22BWizardLM 2 8x22BGPT-4o
CapabilityShared benchmarks50%344349
Price25%276219
Inputs & features15%25070
Context window10%122232
Overall100%29/10039/10043/100
02 — Side by side

Every spec in one table

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

Mixtral 8x22B vs WizardLM 2 8x22B vs GPT-4o specifications side by side
SpecificationMixtral 8x22BMistral AIWizardLM 2 8x22BMicrosoftGPT-4oOpenAI
Capability
Capabilities Index (ECI)122.0—128.1 (best)
ECI rank#149 of 158—#139 of 158 (best)
GPQA DiamondGraduate-level science questions34.1%43.4%48.9% (best)
OTIS Mock AIME 2024–2025Competition mathematics——6.3%
Price per million tokens
Input$2.00$0.493 (best)$2.50
Output$6.00$0.493 (best)$10.00
Cached input——$1.25
Blended (3:1)$3.00$0.493 (best)$4.38
Long-context rateSame rateSame rateSame rate
Price sourceOfficial Mistral APIMedian of 3 providersOfficial OpenAI API
Limits
Context window64,000 tokens65,536 tokens128,000 tokens (best)
Max output64,000 tokens (best)8,000 tokens16,384 tokens
Inputs and features
TextYesYesYes
ImagesNoNoYes
PDFsNoNoYes
AudioNoNoNo
VideoNoNoNo
ReasoningNoNoNo
Tool callingYesNoYes
Structured outputNoNoYes
Availability
WeightsOpenOpenApache-2.0Proprietary
API model IDopen-mixtral-8x22b—gpt-4o
API providers1327 (best)
ReleasedApr 17, 2024Apr 15, 2024May 13, 2024
Knowledge cutoffApr 2024—Sep 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.

  • Mixtral 8x22B$32.00
  • WizardLM 2 8x22B$5.92
  • GPT-4o$45.00
04 — Questions

Which should you choose?

Which is better: Mixtral 8x22B, WizardLM 2 8x22B or GPT-4o?

GPT-4o is the better all-round choice, scoring 43/100 against WizardLM 2 8x22B (39) and Mixtral 8x22B (29). It leads on capability, inputs & features and context window. WizardLM 2 8x22B wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%. Capability uses the average of shared benchmarks (GPQA Diamond), because WizardLM 2 8x22B has no Capabilities Index score yet.

Which is cheaper, Mixtral 8x22B, WizardLM 2 8x22B or GPT-4o?

WizardLM 2 8x22B is cheaper at $0.493 input / $0.493 output per million tokens (median across 3 API providers). Mixtral 8x22B costs $2.00 input / $6.00 output per million tokens (official Mistral API price); GPT-4o costs $2.50 input / $10.00 output per million tokens (official OpenAI API price). At a typical mix of three input tokens to one output token, that is $0.493 per million tokens for WizardLM 2 8x22B versus $3.00 for Mixtral 8x22B (6.1× as much) and $4.38 for GPT-4o (8.9× as much).

Which scores higher on benchmarks?

Not every model here has a Capabilities Index score, so we compare the average of shared benchmarks (GPQA Diamond): GPT-4o 48.9%, WizardLM 2 8x22B 43.4% and Mixtral 8x22B 34.1%. On individual benchmarks: GPQA Diamond — GPT-4o 48.9%, WizardLM 2 8x22B 43.4%, Mixtral 8x22B 34.1%.

Which is better for coding?

There are no published SWE-bench Verified results for Mixtral 8x22B, WizardLM 2 8x22B and GPT-4o yet, so there is no like-for-like coding score. On overall capability, GPT-4o leads, which tends to carry over to coding, but test on your own codebase. Note that WizardLM 2 8x22B does not support tool calling, which most coding agents need.

Which has the bigger context window?

GPT-4o has the largest context window at 128,000 tokens, against 65,536 for WizardLM 2 8x22B and 64,000 for Mixtral 8x22B. Maximum output per response: Mixtral 8x22B up to 64,000, WizardLM 2 8x22B up to 8,000, GPT-4o up to 16,384 tokens.

Which can read images, PDFs, audio or video?

Mixtral 8x22B accepts text; WizardLM 2 8x22B accepts text; GPT-4o accepts text, images and PDFs. GPT-4o handles the widest range of inputs.

Are any of these open source?

Mixtral 8x22B and WizardLM 2 8x22B publishes its weights (Apache-2.0) and can be self-hosted; GPT-4o is proprietary.

Which is newer?

GPT-4o is the newest, released May 13, 2024. Mixtral 8x22B came out Apr 17, 2024; WizardLM 2 8x22B came out Apr 15, 2024. Knowledge cutoff: Mixtral 8x22B Apr 2024, GPT-4o Sep 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.