ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Devstral 2 vs Nova 2 Lite vs GPT-5.1 Codex mini

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

Price in rating: Included · Exclude price

  1. Mistral AI

    Devstral 2

    Released Dec 9, 2025Deprecated

    42/100
    • ECI—
    • Price$0.40 / $2.00
    • Context262K
  2. Amazon

    Nova 2 Lite

    Released Dec 2, 2025

    62/100
    • ECI—
    • Price$0.315 / $2.63
    • Context1M
  3. OpenAI

    GPT-5.1 Codex mini

    Released Nov 13, 2025

    56/100
    • ECI—
    • Price$0.25 / $2.00
    • Context400K
01 — Verdict

Capability evidence is incomplete

There is not enough shared capability evidence to name an overall winner. The scores (Nova 2 Lite 62/100, GPT-5.1 Codex mini 56/100, Devstral 2 42/100) compare the available specifications and 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 priceGPT-5.1 Codex miniGPT-5.1 Codex mini $0.688 · Devstral 2 $0.80 · Nova 2 Lite $0.892 per 1M tokens (3:1 blend)
  • Longest contextNova 2 LiteNova 2 Lite 1,000,000 · GPT-5.1 Codex mini 400,000 · Devstral 2 262,144 tokens
  • Widest inputsNova 2 LiteDevstral 2: Text · Nova 2 Lite: Text, Images, PDFs, Video · GPT-5.1 Codex mini: Text, Images
  • Self-hostingDevstral 2Publishes downloadable weights
How the score is built
MeasureWeightDevstral 2Nova 2 LiteGPT-5.1 Codex mini
Price50%525055
Inputs & features30%258060
Context window20%436450
Specs & price only100%42/10062/10056/100

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.

Devstral 2 vs Nova 2 Lite vs GPT-5.1 Codex mini specifications side by side
SpecificationDevstral 2Mistral AINova 2 LiteAmazonGPT-5.1 Codex miniOpenAI
Capability
Capabilities Index (ECI)———
ECI rank———
Price per million tokens
Input$0.40$0.315$0.25 (best)
Output$2.00 (best)$2.63$2.00 (best)
Cached input———
Blended (3:1)$0.80$0.892$0.688 (best)
Long-context rateSame rateSame rateSame rate
Price sourceOfficial Mistral APIMedian of 6 providersMedian of 16 providers
Limits
Context window262,144 tokens1,000,000 tokens (best)400,000 tokens
Max output262,144 tokens (best)65,536 tokens128,000 tokens
Inputs and features
TextYesYesYes
ImagesNoYesYes
PDFsNoYesNo
AudioNoNoNo
VideoNoYesNo
ReasoningNoYeslow · medium · highYes
Tool callingYesYesYes
Structured outputNoNoNo
Availability
WeightsOpenProprietaryProprietary
API model IDdevstral-2512nova-2-lite-v1—
API providers14816 (best)
ReleasedDec 9, 2025Dec 2, 2025Nov 13, 2025
Knowledge cutoffDec 2025Oct 2025Sep 30, 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.

  • Devstral 2$8.00
  • Nova 2 Lite$8.40
  • GPT-5.1 Codex mini$6.50
04 — Questions

Which should you choose?

Which is better: Devstral 2, Nova 2 Lite or GPT-5.1 Codex mini?

There is not enough shared capability evidence to name an overall winner. The scores (Nova 2 Lite 62/100, GPT-5.1 Codex mini 56/100, Devstral 2 42/100) compare the available specifications and 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, Devstral 2, Nova 2 Lite or GPT-5.1 Codex mini?

GPT-5.1 Codex mini is cheaper at $0.25 input / $2.00 output per million tokens (median across 16 API providers). Devstral 2 costs $0.40 input / $2.00 output per million tokens (official Mistral API price); Nova 2 Lite costs $0.315 input / $2.63 output per million tokens (median across 6 API providers; free on Nova). At a typical mix of three input tokens to one output token, that is $0.688 per million tokens for GPT-5.1 Codex mini versus $0.80 for Devstral 2 (1.2× as much) and $0.892 for Nova 2 Lite (1.3× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Devstral 2 has not been scored yet, Nova 2 Lite has not been scored yet and GPT-5.1 Codex mini has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Devstral 2, Nova 2 Lite and GPT-5.1 Codex mini 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?

Nova 2 Lite has the largest context window at 1,000,000 tokens, against 400,000 for GPT-5.1 Codex mini and 262,144 for Devstral 2. Maximum output per response: Devstral 2 up to 262,144, Nova 2 Lite up to 65,536, GPT-5.1 Codex mini up to 128,000 tokens.

Which can read images, PDFs, audio or video?

Devstral 2 accepts text; Nova 2 Lite accepts text, images, PDFs and video; GPT-5.1 Codex mini accepts text and images. Nova 2 Lite handles the widest range of inputs.

Are any of these open source?

Devstral 2 publishes its weights and can be self-hosted; Nova 2 Lite and GPT-5.1 Codex mini is proprietary.

Which is newer?

Devstral 2 is the newest, released Dec 9, 2025. Nova 2 Lite came out Dec 2, 2025; GPT-5.1 Codex mini came out Nov 13, 2025. Knowledge cutoff: Devstral 2 Dec 2025, Nova 2 Lite Oct 2025, GPT-5.1 Codex mini Sep 30, 2024.

How do you decide the winner?

This rating uses price (50%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (30%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (20%, 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.