Which AI model should I use today?
Checked Nov 14, 2026
Sonnet 5.5Pick
$2 / $10
Solves 44% of Terminal-Bench 4.0 tasks at high effort, at about two-thirds of Opus's cost per task.
model: "claude-sonnet-5-5", output_config: { effort: "high" }Opus 5.5
$4 / $20
Twice the token price. 53% at medium effort, about 1.5x Sonnet's cost per task. Good for planning architecture.
Fable 5.2
$8 / $40
Four times Sonnet's token price. Keep it for a bug that Sonnet and Opus both failed to fix.
Haiku 5.5
$0.10 / $0.50
A cheap subagent for side jobs. Keep a stronger model as your main coder.
Sonnet 5Older
$2 / $10
The previous generation, at the same price, so there's no reason to stay on it.
Medium effort is enough for small, well-scoped edits.
Prices are per million tokens, input / output. We pick for the best result per dollar, so a pricier model can score higher and still lose out. This pick last changed Oct 9, 2026. Sources: Anthropic docs, OpenAI docs, Gemini docs, Artificial Analysis, Vals AI, Terminal-Bench.
![]()
Maintained by the AI team at Vegan Hacktivists
| Lab | Task | Model | Effort | Why |
|---|---|---|---|---|
| Claude | Code | Sonnet 5.5 claude-sonnet-5-5 | High | Solves 44% of Terminal-Bench 4.0 tasks at high effort, at about two-thirds of Opus's cost per task. |
| Claude | Everyday | Opus 5.5 claude-opus-5-5 | Low | Anthropic's suggested starting model. Scores 35 on the Artificial Analysis index at low effort, in about 3 minutes a task. |
| Claude | Research | Fable 5.2 claude-fable-5-2 | High | Tops the Artificial Analysis Intelligence Index at high effort (62), ten points ahead of Opus 5.5 at high. |
| Claude | Agents | Sonnet 5.5 claude-sonnet-5-5 | Medium | Leads Opus on the Vals Index (70.9% vs 70.1%, measured at max effort) at two-thirds of the cost per task. |
| OpenAI | Code | GPT-6.1 Sol gpt-6.1-sol | High | Solves 51.5% of Terminal-Bench 4.0 tasks at high effort, 2.5 points behind Astra, at a fifth of the token price. |
| OpenAI | Everyday | GPT-6.1 Sol gpt-6.1-sol | Low | At low effort it finishes tasks in about 2 minutes on Artificial Analysis, half Astra's time at low. |
| OpenAI | Research | GPT-6 Astra gpt-6-astra | High | Tops OpenAI's lineup on the Artificial Analysis index at high effort (55, against 50 for 6.1 Sol). |
| OpenAI | Agents | GPT-6.1 Sol gpt-6.1-sol | High | Scores 64.5% on AutomationBench-AA at high effort, 2.1 points under Astra, at under a fifth of the cost per task. |
| Gemini | Code | Carbon gemini-4-carbon | High | At high effort, scores 60 vs. Argon's 50 on the Artificial Analysis index, with about 10% higher average benchmark cost. |
| Gemini | Everyday | Hydrogen gemini-4-hydrogen | High | Scores 51 on the Artificial Analysis index at high effort, above Argon's 50, and finishes tasks about 3.5x faster. |
| Gemini | Research | Carbon gemini-4-carbon | High | Leads Google's lineup on the Artificial Analysis Intelligence Index at high effort (60 vs 50 for Argon). |
| Gemini | Agents | Carbon gemini-4-carbon | Medium | Scores 58 on the Artificial Analysis index at medium effort, above Argon's 50 at high, for less per task. |
Want helpputting this to work?
Vegan Hacktivists offers free AI and automation support to animal protection organizations — from a first conversation to a full build.

