ZMIME
Comparison · 3 models · Updated Nov 2, 2026

Codestral 25.08 vs ERNIE 4.5 VL 424B A47B vs Qwen3 Coder Flash

Qwen3 Coder Flash comes out ahead, 51 to 47 and 44 on our weighted score, though Codestral 25.08 is 25% cheaper per token.

  1. Mistral AI

    Codestral 25.08

    Released Jul 30, 2025

    47/100
    • ECI—
    • Price$0.30 / $0.90
    • Context128K
    Full specs Remove Codestral 25.08

  2. Baidu

    ERNIE 4.5 VL 424B A47B

    Released Jun 30, 2025

    44/100
    • ECI—
    • Price$0.42 / $1.25
    • Context123K
    Full specs Remove ERNIE 4.5 VL 424B A47B
  3. Our pick

    Alibaba (Qwen)

    Qwen3 Coder Flash

    Released Jul 28, 2025

    51/100
    • ECI—
    • Price$0.30 / $1.50
    • Context1M
    Full specs Remove Qwen3 Coder Flash
01 — Verdict

Qwen3 Coder Flash is our pick

Qwen3 Coder Flash is the better all-round choice, scoring 51/100 against Codestral 25.08 (47) and ERNIE 4.5 VL 424B A47B (44). It leads on context window. Codestral 25.08 wins on price. 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 priceCodestral 25.08Codestral 25.08 $0.45 · Qwen3 Coder Flash $0.60 · ERNIE 4.5 VL 424B A47B $0.627 per 1M tokens (3:1 blend)
  • Longest contextQwen3 Coder FlashQwen3 Coder Flash 1,000,000 · Codestral 25.08 128,000 · ERNIE 4.5 VL 424B A47B 123,000 tokens
  • Widest inputsERNIE 4.5 VL 424B A47B Codestral 25.08: Text · ERNIE 4.5 VL 424B A47B : Text, Images · Qwen3 Coder Flash: Text
  • Self-hostingERNIE 4.5 VL 424B A47B Publishes downloadable weights
How the score is built
MeasureWeightCodestral 25.08ERNIE 4.5 VL 424B A47B Qwen3 Coder Flash
Price50%645758
Inputs & features30%353535
Context window20%242360
Overall100%47/10044/10051/100

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.

Codestral 25.08 vs ERNIE 4.5 VL 424B A47B vs Qwen3 Coder Flash specifications side by side
SpecificationCodestral 25.08Mistral AIERNIE 4.5 VL 424B A47B BaiduQwen3 Coder FlashAlibaba (Qwen)
Capability
Capabilities Index (ECI)———
ECI rank———
Price per million tokens
Input$0.30 (best)$0.42$0.30 (best)
Output$0.90 (best)$1.25$1.50
Cached input$0.03 (best)—$0.06
Blended (3:1)$0.45 (best)$0.627$0.60
Long-context rateSame rateSame rateOver 32K: $0.50 / $2.50
Price sourceOfficial Mistral APIMedian of 1 providersOfficial Alibaba API
Limits
Context window128,000 tokens123,000 tokens1,000,000 tokens (best)
Max output8,192 tokens16,000 tokens65,536 tokens (best)
Inputs and features
TextYesYesYes
ImagesNoYesNo
PDFsNoNoNo
AudioNoNoNo
VideoNoNoNo
ReasoningNoYesNo
Tool callingYesNoYes
Structured outputYesNoYes
Availability
WeightsProprietaryOpenProprietary
API model IDcodestral-2508—qwen3-coder-flash
API providers11 (best)110
ReleasedJul 30, 2025Jun 30, 2025Jul 28, 2025
Knowledge cutoff—Mar 31, 2025Apr 2025
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.

  • Codestral 25.08$4.80
  • ERNIE 4.5 VL 424B A47B $6.70
  • Qwen3 Coder Flash$6.00
04 — Questions

Which should you choose?

Which is better: Codestral 25.08, ERNIE 4.5 VL 424B A47B or Qwen3 Coder Flash?

Qwen3 Coder Flash is the better all-round choice, scoring 51/100 against Codestral 25.08 (47) and ERNIE 4.5 VL 424B A47B (44). It leads on context window. Codestral 25.08 wins on price. 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, Codestral 25.08, ERNIE 4.5 VL 424B A47B or Qwen3 Coder Flash?

Codestral 25.08 is cheaper at $0.30 input / $0.90 output per million tokens (official Mistral API price). Qwen3 Coder Flash costs $0.30 input / $1.50 output per million tokens (official Alibaba API price); ERNIE 4.5 VL 424B A47B costs $0.42 input / $1.25 output per million tokens (median across 1 API provider). At a typical mix of three input tokens to one output token, that is $0.45 per million tokens for Codestral 25.08 versus $0.60 for Qwen3 Coder Flash (1.3× as much) and $0.627 for ERNIE 4.5 VL 424B A47B (1.4× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Codestral 25.08 has not been scored yet, ERNIE 4.5 VL 424B A47B has not been scored yet and Qwen3 Coder Flash has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Codestral 25.08, ERNIE 4.5 VL 424B A47B and Qwen3 Coder Flash yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Note that ERNIE 4.5 VL 424B A47B does not support tool calling, which most coding agents need.

Which has the bigger context window?

Qwen3 Coder Flash has the largest context window at 1,000,000 tokens, against 128,000 for Codestral 25.08 and 123,000 for ERNIE 4.5 VL 424B A47B . Maximum output per response: Codestral 25.08 up to 8,192, ERNIE 4.5 VL 424B A47B up to 16,000, Qwen3 Coder Flash up to 65,536 tokens.

Which can read images, PDFs, audio or video?

Codestral 25.08 accepts text; ERNIE 4.5 VL 424B A47B accepts text and images; Qwen3 Coder Flash accepts text. ERNIE 4.5 VL 424B A47B handles the widest range of inputs.

Are any of these open source?

ERNIE 4.5 VL 424B A47B publishes its weights and can be self-hosted; Codestral 25.08 and Qwen3 Coder Flash is proprietary.

Which is newer?

Codestral 25.08 is the newest, released Jul 30, 2025. Qwen3 Coder Flash came out Jul 28, 2025; ERNIE 4.5 VL 424B A47B came out Jun 30, 2025. Knowledge cutoff: ERNIE 4.5 VL 424B A47B Mar 31, 2025, Qwen3 Coder Flash Apr 2025.

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