Gemma 3 4B vs gpt-oss-120b vs Grok 4.20
Gemma 3 4B at $0.058 in and $0.115 out, gpt-oss-120b at $0.043 in and $0.196 out, and Grok 4.20 at $1.44 in and $2.88 out — per million tokens, from the same wallet.
Gemma 3 4B
0.04¢
for a client proposal
$0.058 in · $0.115 out / 1M
131K context · released March 2025
gpt-oss-120b
OpenAI
0.04¢
for a client proposal
$0.043 in · $0.196 out / 1M
131K context · released August 2025
Grok 4.20
xAI
0.95¢
for a client proposal
$1.44 in · $2.88 out / 1M
2M context · released 4 months ago
Asking all three at once — identical prompt, identical context — costs about 1¢ for a typical client proposal.
At a glance
- Cheapest input: gpt-oss-120b at $0.043 per million tokens.
- Cheapest output: Gemma 3 4B at $0.115 per million tokens.
- Largest context: Grok 4.20 at 2M tokens — about 2,900 pages.
Cost per task
Estimated totals at the rates above. Solid bar is input, lighter bar is output.
Quick code fix
2K in · 500 outDraft a client proposal from notes
5K in · 800 outContext-heavy session (Alyph workspace)
155K in · 4K outAnalyze a 50-page legal contract
35K in · 1K outLarge architecture refactor
250K in · 3K outLong-context rate applied at this input size.
Solid part of each bar is input tokens, the lighter part is output. Models that don’t fit a task are marked instead of priced.
Cost vs Token Amount
Drag the slider to adjust the split between input and output workload.
Adjust the ratio slider to change how output tokens influence the final price for equivalent workloads.
Pricing
Specs
Try it with real numbers
2K tokens in · 500 tokens out.
All 3, same prompt and context: 0.47¢. That is the entire cost of the comparison.
Your $5 welcome credit covers about 1,071 of these.
Which should you pick?
For long documents, Grok 4.20 has the largest context window here — 2M tokens.
On price, gpt-oss-120b is the cheapest of the three on a typical task, at about 0.04¢.
For images or files, Gemma 3 4B and Grok 4.20 can read them directly; gpt-oss-120b is text-only.
If you want to watch the model think, gpt-oss-120b and Grok 4.20 show reasoning; the other answers directly.
Or don’t pick. Send the identical prompt to all three, read the answers side by side, and keep the winner. How to run a fair bake-off →
Questions, answered
Which is cheapest — Gemma 3 4B, gpt-oss-120b, or Grok 4.20?
Which has the largest context window?
Can I run Gemma 3 4B, gpt-oss-120b, and Grok 4.20 side by side on Alyph?
Ask all three at once
Identical prompt, identical context, answers side by side — about 1¢ for a typical client proposal. Free to start, with $5 of credit.