ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Gemini 2.5 Computer Use Preview vs GPT-5-Codex vs Rnj-1 Instruct

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

Price in rating: Excluded · Include price

  1. Google

    Gemini 2.5 Computer Use Preview

    Released Oct 7, 2025

    49/100
    • ECI—
    • Price$1.25 / $10.00
    • Context128K
  2. OpenAI

    GPT-5-Codex

    Released Sep 15, 2025

    56/100
    • ECI—
    • Price$1.25 / $10.00
    • Context400K
  3. Essential AI

    Rnj-1 Instruct

    Released Dec 4, 2025

    19/100
    • ECI—
    • Price$0.15 / $0.15
    • Context33K
01 — Verdict

Capability evidence is incomplete

There is not enough shared capability evidence to name an overall winner. The scores (GPT-5-Codex 56/100, Gemini 2.5 Computer Use Preview 49/100, Rnj-1 Instruct 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 priceRnj-1 InstructRnj-1 Instruct $0.15 · Gemini 2.5 Computer Use Preview $3.44 · GPT-5-Codex $3.44 per 1M tokens (3:1 blend)
  • Longest contextGPT-5-CodexGPT-5-Codex 400,000 · Gemini 2.5 Computer Use Preview 128,000 · Rnj-1 Instruct 32,768 tokens
  • Widest inputsGemini 2.5 Computer Use Preview and GPT-5-CodexGemini 2.5 Computer Use Preview: Text, Images · GPT-5-Codex: Text, Images · Rnj-1 Instruct: Text
  • Self-hostingRnj-1 InstructPublishes downloadable weights (Apache 2.0)
How the score is built
MeasureWeightGemini 2.5 Computer Use PreviewGPT-5-CodexRnj-1 Instruct
Inputs & features60%606025
Context window40%325011
Specs only100%49/10056/10019/100

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.

Gemini 2.5 Computer Use Preview vs GPT-5-Codex vs Rnj-1 Instruct specifications side by side
SpecificationGemini 2.5 Computer Use PreviewGoogleGPT-5-CodexOpenAIRnj-1 InstructEssential AI
Capability
Capabilities Index (ECI)———
ECI rank———
Price per million tokens
Input$1.25$1.25$0.15 (best)
Output$10.00$10.00$0.15 (best)
Cached input———
Blended (3:1)$3.44$3.44$0.15 (best)
Long-context rateSame rateSame rateSame rate
Price sourceOfficial Google APIMedian of 9 providersMedian of 1 providers
Limits
Context window128,000 tokens400,000 tokens (best)32,768 tokens
Max output64,000 tokens128,000 tokens (best)32,768 tokens
Inputs and features
TextYesYesYes
ImagesYesYesNo
PDFsNoNoNo
AudioNoNoNo
VideoNoNoNo
ReasoningYesYesNo
Tool callingYesYesYes
Structured outputNoNoNo
Availability
WeightsProprietaryProprietaryOpenApache 2.0
API model IDgemini-2.5-computer-use-preview-10-2025——
API providers210 (best)1
ReleasedOct 7, 2025Sep 15, 2025Dec 4, 2025
Knowledge cutoffJan 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.

  • Gemini 2.5 Computer Use Preview$32.50
  • GPT-5-Codex$32.50
  • Rnj-1 Instruct$1.80
04 — Questions

Which should you choose?

Which is better: Gemini 2.5 Computer Use Preview, GPT-5-Codex or Rnj-1 Instruct?

There is not enough shared capability evidence to name an overall winner. The scores (GPT-5-Codex 56/100, Gemini 2.5 Computer Use Preview 49/100, Rnj-1 Instruct 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, Gemini 2.5 Computer Use Preview, GPT-5-Codex or Rnj-1 Instruct?

Rnj-1 Instruct is cheaper at $0.15 input / $0.15 output per million tokens (median across 1 API provider). Gemini 2.5 Computer Use Preview costs $1.25 input / $10.00 output per million tokens (official Google API price); GPT-5-Codex costs $1.25 input / $10.00 output per million tokens (median across 9 API providers). At a typical mix of three input tokens to one output token, that is $0.15 per million tokens for Rnj-1 Instruct versus $3.44 for Gemini 2.5 Computer Use Preview (23× as much) and $3.44 for GPT-5-Codex (23× as much).

Which scores higher on benchmarks?

There is no independent benchmark that covers all three models yet. Gemini 2.5 Computer Use Preview has not been scored yet, GPT-5-Codex has not been scored yet and Rnj-1 Instruct has not been scored yet.

Which is better for coding?

There are no published SWE-bench Verified results for Gemini 2.5 Computer Use Preview, GPT-5-Codex and Rnj-1 Instruct 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?

GPT-5-Codex has the largest context window at 400,000 tokens, against 128,000 for Gemini 2.5 Computer Use Preview and 32,768 for Rnj-1 Instruct. Maximum output per response: Gemini 2.5 Computer Use Preview up to 64,000, GPT-5-Codex up to 128,000, Rnj-1 Instruct up to 32,768 tokens.

Which can read images, PDFs, audio or video?

Gemini 2.5 Computer Use Preview accepts text and images; GPT-5-Codex accepts text and images; Rnj-1 Instruct accepts text. Gemini 2.5 Computer Use Preview handles the widest range of inputs.

Are any of these open source?

Rnj-1 Instruct publishes its weights (Apache 2.0) and can be self-hosted; Gemini 2.5 Computer Use Preview and GPT-5-Codex is proprietary.

Which is newer?

Rnj-1 Instruct is the newest, released Dec 4, 2025. Gemini 2.5 Computer Use Preview came out Oct 7, 2025; GPT-5-Codex came out Sep 15, 2025. Knowledge cutoff: Gemini 2.5 Computer Use Preview Jan 2025, GPT-5-Codex Sep 30, 2024.

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.