ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Baichuan M2 32B vs Codestral 25.08 vs Qwen3 VL Flash (Alibaba Cloud)

No independent capability benchmark covers every model here yet. The score uses inputs & features and context window.

Price in rating: Excluded · Include price

  1. Baichuan

    Baichuan M2 32B

    Released Aug 13, 2025

    19/100
    • ECI—
    • Price$0.07 / $0.07
    • Context131K
  2. Mistral AI

    Codestral 25.08

    Released Jul 30, 2025

    34/100
    • ECI—
    • Price$0.30 / $0.90
    • Context128K
  3. Alibaba (Qwen)

    Qwen3 VL Flash (Alibaba Cloud)

    Released Oct 15, 2025

    47/100
    • ECI—
    • Price$0.022 / $0.215
    • Context262K
01 — Verdict

Capability evidence is incomplete

There is not enough shared capability evidence to name an overall winner. The scores (Qwen3 VL Flash (Alibaba Cloud) 47/100, Codestral 25.08 34/100, Baichuan M2 32B 19/100) compare the available specifications only. The score weighs inputs & features 60%, context window 40%. 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 priceBaichuan M2 32BBaichuan M2 32B $0.07 · Qwen3 VL Flash (Alibaba Cloud) $0.07 · Codestral 25.08 $0.45 per 1M tokens (3:1 blend)
  • Longest contextQwen3 VL Flash (Alibaba Cloud)Qwen3 VL Flash (Alibaba Cloud) 262,144 · Baichuan M2 32B 131,072 · Codestral 25.08 128,000 tokens
  • Widest inputsQwen3 VL Flash (Alibaba Cloud)Baichuan M2 32B: Text · Codestral 25.08: Text · Qwen3 VL Flash (Alibaba Cloud): Text, Images
  • Self-hostingBaichuan M2 32BPublishes downloadable weights (Apache-2.0)
How the score is built
MeasureWeightBaichuan M2 32BCodestral 25.08Qwen3 VL Flash (Alibaba Cloud)
Inputs & features60%103550
Context window40%323243
Specs only100%19/10034/10047/100

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.

Baichuan M2 32B vs Codestral 25.08 vs Qwen3 VL Flash (Alibaba Cloud) specifications side by side
SpecificationBaichuan M2 32BBaichuanCodestral 25.08Mistral AIQwen3 VL Flash (Alibaba Cloud)Alibaba (Qwen)
Capability
Capabilities Index (ECI)———
ECI rank———
Price per million tokens
Input$0.07$0.30$0.022 (best)
Output$0.07 (best)$0.90$0.215
Cached input—$0.03—
Blended (3:1)$0.07 (best)$0.45$0.07
Long-context rateSame rateSame rateSame rate
Price sourceMedian of 1 providersOfficial Mistral APIMedian of 1 providers
Limits
Context window131,072 tokens128,000 tokens262,144 tokens (best)
Max output131,072 tokens (best)8,192 tokens32,768 tokens
Inputs and features
TextYesYesYes
ImagesNoNoYes
PDFsNoNoNo
AudioNoNoNo
VideoNoNoNo
ReasoningYesNoNo
Tool callingNoYesYes
Structured outputNoYesNo
Availability
WeightsOpenApache-2.0ProprietaryProprietary
API model ID—codestral-2508—
API providers111 (best)1
ReleasedAug 13, 2025Jul 30, 2025Oct 15, 2025
Knowledge cutoff———
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.

  • Baichuan M2 32B$0.84
  • Codestral 25.08$4.80
  • Qwen3 VL Flash (Alibaba Cloud)$0.65
04 — Questions

Which should you choose?

Which is better: Baichuan M2 32B, Codestral 25.08 or Qwen3 VL Flash (Alibaba Cloud)?

There is not enough shared capability evidence to name an overall winner. The scores (Qwen3 VL Flash (Alibaba Cloud) 47/100, Codestral 25.08 34/100, Baichuan M2 32B 19/100) compare the available specifications only. The score weighs inputs & features 60%, context window 40%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.

Which is cheaper, Baichuan M2 32B, Codestral 25.08 or Qwen3 VL Flash (Alibaba Cloud)?

Baichuan M2 32B is cheaper at $0.07 input / $0.07 output per million tokens (median across 1 API provider). Qwen3 VL Flash (Alibaba Cloud) costs $0.022 input / $0.215 output per million tokens (median across 1 API provider); Codestral 25.08 costs $0.30 input / $0.90 output per million tokens (official Mistral API price). At a typical mix of three input tokens to one output token, that is $0.07 per million tokens for Baichuan M2 32B versus $0.07 for Qwen3 VL Flash (Alibaba Cloud) (1× as much) and $0.45 for Codestral 25.08 (6.4× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Baichuan M2 32B has not been scored yet, Codestral 25.08 has not been scored yet and Qwen3 VL Flash (Alibaba Cloud) has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Baichuan M2 32B, Codestral 25.08 and Qwen3 VL Flash (Alibaba Cloud) yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Note that Baichuan M2 32B does not support tool calling, which most coding agents need.

Which has the bigger context window?

Qwen3 VL Flash (Alibaba Cloud) has the largest context window at 262,144 tokens, against 131,072 for Baichuan M2 32B and 128,000 for Codestral 25.08. Maximum output per response: Baichuan M2 32B up to 131,072, Codestral 25.08 up to 8,192, Qwen3 VL Flash (Alibaba Cloud) up to 32,768 tokens.

Which can read images, PDFs, audio or video?

Baichuan M2 32B accepts text; Codestral 25.08 accepts text; Qwen3 VL Flash (Alibaba Cloud) accepts text and images. Qwen3 VL Flash (Alibaba Cloud) handles the widest range of inputs.

Are any of these open source?

Baichuan M2 32B publishes its weights (Apache-2.0) and can be self-hosted; Codestral 25.08 and Qwen3 VL Flash (Alibaba Cloud) is proprietary.

Which is newer?

Qwen3 VL Flash (Alibaba Cloud) is the newest, released Oct 15, 2025. Baichuan M2 32B came out Aug 13, 2025; Codestral 25.08 came out Jul 30, 2025.

How do you decide the winner?

This rating uses inputs & features (60%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (40%, 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.