Download

Gemini 4 Argon vs GPT-6.1 Sol

Gemini 4 Argon and GPT-6.1 Sol match on price and context today: $2 input, $10 output and $0.10 cached input per 1M tokens, and 1.05M tokens each. Two things split them. The Argon price is introductory and rises to $4 and $20 on 1 December 2026, and GPT-6.1 Sol writes replies of up to 128K tokens against 66K for Argon.

Identical sessions until December

With every rate the same, our sessions cost the same on both: $0.65 for a quick fix, $10.80 for the feature session, $97.50 for a long agent day, and $2.59 for the feature session with 90% of prompts cached. Filling the whole window once is $2.10 on each.

From 1 December 2026 the Argon input and output prices double. The price note we cite does not say whether its cached rate changes.

Benchmarks

There is one score between them: Argon's 57.4% on Terminal-Bench 4.0, from Google's comparison table as reproduced by HackerNoon. We found no score for GPT-6.1 Sol that we could cite, so we can't say which is stronger.

Dates and inputs

GPT-6.1 Sol was first listed on the API on 29 September 2026, and Google released Argon the next day, 30 September. Argon also reads video, audio and PDF files; GPT-6.1 Sol reads text and images. Against Anthropic's model at the same price, see Sonnet 5.5 against Gemini 4 Argon.

Side by side

Gemini 4 Argon and GPT-6.1 Sol: specifications and prices
Gemini 4 ArgonGPT-6.1 Sol
MakerGoogleOpenAI
Released30 September 2026[1]29 September 2026[2] first API listing
Context window1.05M tokens[3]1.05M tokens[4]
Max output66K tokens[3]128K tokens[4]
Input price$2 per 1M tokens[5]$2 per 1M tokens[4]
Output price$10 per 1M tokens[5]$10 per 1M tokens[4]
Cached input$0.10 per 1M tokens[5]$0.10 per 1M tokens[4]
Long promptsOne priceOne price
Price changeIntroductory price through 30 November 2026; $4 input and $20 output per 1M tokens from 1 December 2026.[5]None announced
Open weightsNo[6] Served through the Gemini API; no weights are published.No[7] Listed as an API model, not in OpenAI's open-weight section.
Readstext, image, video, audio, pdf[3]text, image[4]

Benchmarks

None of the benchmarks below has a published score for both models. We only show a score we can cite with its source and date; "Not published" means we found none for that model, not that it scored zero.

Published benchmark scores (higher is better)
Gemini 4 ArgonGPT-6.1 Sol
Terminal-Bench 4.0Gemini 4 Argon: 57.4%57.4%GPT-6.1 Sol: not published
Benchmarks: Gemini 4 Argon and GPT-6.1 Sol
BenchmarkGemini 4 ArgonGPT-6.1 Sol
Terminal-Bench 4.0Agentic coding tasks in a real terminal.57.4%[8] Google's comparison table as reproduced by HackerNoon.Not published
FrontierCode 1.1 (Main)Hard software engineering tasks.Not publishedNot published
Humanity's Last Exam (with tools)Expert-level questions across many fields, tools allowed.Not publishedNot published

What a coding session costs

Our own estimate from the list prices above, for three sessions we define: a coding agent resends the conversation with every call, so prompts are large and grow. "Cached" assumes 90% of prompt tokens are read from the prompt cache and leaves out cache write fees. A long-prompt price applies when a call's prompt is over its threshold. Work out your own mix with the token cost calculator.

Feature session at list prices, US dollars (lower is cheaper)
Gemini 4 ArgonGPT-6.1 Sol
No cachingGemini 4 Argon: $10.80$10.80GPT-6.1 Sol: $10.80$10.8090% of prompt tokens from cacheGemini 4 Argon: $2.59$2.59GPT-6.1 Sol: $2.59$2.59
Session costs: Gemini 4 Argon and GPT-6.1 Sol
SessionGemini 4 ArgonGemini 4 Argon cachedGPT-6.1 SolGPT-6.1 Sol cached
Quick fix10 calls of 25K prompt and 1.5K output tokens$0.65$0.22$0.65$0.22
Feature session60 calls of 80K prompt and 2K output tokens$10.80$2.59$10.80$2.59
Long agent day300 calls of 150K prompt and 2.5K output tokens$97.50$20.55$97.50$20.55

Value for money and release gap

Gemini 4 Argon is dated 1 day later (GPT-6.1 Sol: first API listing). Price per point divides our feature-session cost by a published score: lower means more score for the money. It is a rough lens, not a ranking.

Value for money: Gemini 4 Argon and GPT-6.1 Sol
MeasureGemini 4 ArgonGPT-6.1 Sol
Input tokens per $1500K500K
Filling the whole context window once$2.10$2.10

Questions

Do Gemini 4 Argon and GPT-6.1 Sol cost the same?

Until 30 November 2026, yes: $2 per 1M input tokens, $10 per 1M output and $0.10 cached. From 1 December 2026 Gemini 4 Argon costs $4 and $20.

Which is better, Gemini 4 Argon or GPT-6.1 Sol?

We have one benchmark score between them: the 57.4% Gemini 4 Argon scored on Terminal-Bench 4.0. We found no score for GPT-6.1 Sol that we could cite.

Which writes longer replies?

GPT-6.1 Sol, up to 128K output tokens, against 66K for Gemini 4 Argon. Their context windows are the same size, 1.05M tokens.

Most compared

Any other two models: pick them in the lists at the top of the page.

More tools

Sources

  1. blog.google: innovation-and-ai/models-and-research/gemini-models/gemini-4-argon, read
  2. openrouter.ai: openai/gpt-6.1-sol, read
  3. ai.google.dev: gemini-api/docs/models/gemini-4-argon, read
  4. developers.openai.com: api/docs/models/gpt-6.1-sol, read
  5. ai.google.dev: gemini-api/docs/pricing, read
  6. ai.google.dev: gemini-api/docs/models, read
  7. developers.openai.com: api/docs/models/all, read
  8. hackernoon.com: gemini-4-argon-how-does-it-do-against-gpt-6-astra-and-claude, read

Numbers from the makers' own pages, checked by hand, last read . List prices in US dollars before tax. Spotted a mistake? office@shuriken.sh.

Made by Shuriken, the harness for coding agents. Its hot swapping moves your chats to another of your accounts or API keys when one nears its limit. Use only accounts you own and are entitled to use.

Try Shuriken on your own computer

Add your Claude, Codex and Gemini accounts once. Shuriken switches between them so your agents keep working.

Download Pricing
Download on the App StoreGet it on Google PlayComing soon