ZMIME
Comparison · 3 models · Updated Oct 4, 2026

Qwen2.5-VL 7B Instruct vs Qwen3-VL 30B-A3B vs Codestral

Qwen3-VL 30B-A3B comes out ahead, 59 to 51 and 48 on our weighted score, and it is the cheaper option too.

  1. Alibaba (Qwen)

    Qwen2.5-VL 7B Instruct

    Released Sep 2024

    51/100
    • ECI—
    • Price$0.35 / $1.05
    • Context131K
    Full specs Remove Qwen2.5-VL 7B Instruct
  2. Our pick

    Alibaba (Qwen)

    Qwen3-VL 30B-A3B

    Released Apr 2025

    59/100
    • ECI—
    • Price$0.20 / $0.80
    • Context131K
    Full specs Remove Qwen3-VL 30B-A3B

  3. Mistral AI

    Codestral

    Released May 29, 2024

    48/100
    • ECI—
    • Price$0.30 / $0.90
    • Context256K
    Full specs Remove Codestral
01 — Verdict

Qwen3-VL 30B-A3B is our pick

Qwen3-VL 30B-A3B is the better all-round choice, scoring 59/100 against Qwen2.5-VL 7B Instruct (51) and Codestral (48). It leads on price and inputs & features. Codestral wins on context window. 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 priceQwen3-VL 30B-A3BQwen3-VL 30B-A3B $0.35 · Codestral $0.45 · Qwen2.5-VL 7B Instruct $0.525 per 1M tokens (3:1 blend)
  • Longest contextCodestralCodestral 256,000 · Qwen2.5-VL 7B Instruct 131,072 · Qwen3-VL 30B-A3B 131,072 tokens
  • Widest inputsQwen2.5-VL 7B Instruct and Qwen3-VL 30B-A3BQwen2.5-VL 7B Instruct: Text, Images · Qwen3-VL 30B-A3B: Text, Images · Codestral: Text
  • Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
How the score is built
MeasureWeightQwen2.5-VL 7B InstructQwen3-VL 30B-A3BCodestral
Price50%637266
Inputs & features30%506025
Context window20%242436
Overall100%51/10059/10048/100

Left out because at least one model lacks the data: capability. 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.

Qwen2.5-VL 7B Instruct vs Qwen3-VL 30B-A3B vs Codestral specifications side by side
SpecificationQwen2.5-VL 7B InstructAlibaba (Qwen)Qwen3-VL 30B-A3BAlibaba (Qwen)CodestralMistral AI
Capability
Capabilities Index (ECI)———
ECI rank———
Price per million tokens
Input$0.35$0.20 (best)$0.30
Output$1.05$0.80 (best)$0.90
Cached input——$0.03
Blended (3:1)$0.525$0.35 (best)$0.45
Long-context rateSame rateSame rateSame rate
Price sourceOfficial Alibaba APIOfficial Alibaba APIOfficial Mistral API
Limits
Context window131,072 tokens131,072 tokens256,000 tokens (best)
Max output8,192 tokens32,768 tokens (best)4,096 tokens
Inputs and features
TextYesYesYes
ImagesYesYesNo
PDFsNoNoNo
AudioNoNoNo
VideoNoNoNo
ReasoningNoYesNo
Tool callingYesYesYes
Structured outputNoNoNo
Availability
WeightsOpenOpenOpen
API model IDqwen2-5-vl-7b-instructqwen3-vl-30b-a3bcodestral-latest
API providers113 (best)
ReleasedSep 2024Apr 2025May 29, 2024
Knowledge cutoffApr 2024Apr 2025Oct 2024
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.

  • Qwen2.5-VL 7B Instruct$5.60
  • Qwen3-VL 30B-A3B$3.60
  • Codestral$4.80
04 — Questions

Which should you choose?

Which is better: Qwen2.5-VL 7B Instruct, Qwen3-VL 30B-A3B or Codestral?

Qwen3-VL 30B-A3B is the better all-round choice, scoring 59/100 against Qwen2.5-VL 7B Instruct (51) and Codestral (48). It leads on price and inputs & features. Codestral wins on context window. 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, Qwen2.5-VL 7B Instruct, Qwen3-VL 30B-A3B or Codestral?

Qwen3-VL 30B-A3B is cheaper at $0.20 input / $0.80 output per million tokens (official Alibaba API price). Codestral costs $0.30 input / $0.90 output per million tokens (official Mistral API price); Qwen2.5-VL 7B Instruct costs $0.35 input / $1.05 output per million tokens (official Alibaba API price). At a typical mix of three input tokens to one output token, that is $0.35 per million tokens for Qwen3-VL 30B-A3B versus $0.45 for Codestral (1.3× as much) and $0.525 for Qwen2.5-VL 7B Instruct (1.5× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Qwen2.5-VL 7B Instruct has not been scored yet, Qwen3-VL 30B-A3B has not been scored yet and Codestral has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Qwen2.5-VL 7B Instruct, Qwen3-VL 30B-A3B and Codestral 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?

Codestral has the largest context window at 256,000 tokens, against 131,072 for Qwen2.5-VL 7B Instruct and 131,072 for Qwen3-VL 30B-A3B. Maximum output per response: Qwen2.5-VL 7B Instruct up to 8,192, Qwen3-VL 30B-A3B up to 32,768, Codestral up to 4,096 tokens.

Which can read images, PDFs, audio or video?

Qwen2.5-VL 7B Instruct accepts text and images; Qwen3-VL 30B-A3B accepts text and images; Codestral accepts text. Qwen2.5-VL 7B Instruct 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?

Qwen3-VL 30B-A3B is the newest, released Apr 2025. Qwen2.5-VL 7B Instruct came out Sep 2024; Codestral came out May 29, 2024. Knowledge cutoff: Qwen2.5-VL 7B Instruct Apr 2024, Qwen3-VL 30B-A3B Apr 2025, Codestral Oct 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.

05 — Keep comparing

Related comparisons