Open weights
Which model to run yourself, or buy from whoever you like. Four answers — best value for money, best there is, best with open weights, cheapest that still works.
Which model to run yourself, or buy from whoever you like. 117 measured, 100 of them for sale, and 70 of those stand on two or more of these boards — only those can be picked.
- Best value for moneyPhi-3-mini-4k-instructMicrosoft · 14th of 134 on Arc ai2 · 2 boards$0.13→$0.52per Mtok in / out2 sellers
- Best frontierPhi-3-small-8k-instructMicrosoft · 6th of 134 on Arc ai2 · 2 boards$0.15→$0.6per Mtok in / out1 seller
- CheapestQwen2.5-Coder-7BAlibaba · 38th of 134 on Arc ai2 · 2 boards$0.025→$0.05per Mtok in / out3 sellers
Judged on
- Artificial Analysis · Intelligence Index
- Aider polyglot coding leaderboard
- ARC-AGI-1 (Semi-Private)
- ARC-AGI-1 (Public Eval)
- ARC-AGI-2 (Semi-Private)
- ARC-AGI-2 (Public Eval)
- ARC-AGI-3 (Semi-Private)
- Algotune — Epoch AI
- Apex agents — Epoch AI
- Arc agi — Epoch AI
- Arc agi 2 — Epoch AI
- Arc ai2 — Epoch AI
- Blueprint bench 2 — Epoch AI
- Chess puzzles — Epoch AI
- Critpt — Epoch AI
- Cursorbench — Epoch AI
- Enigma eval — Epoch AI
- Frontiercode — Epoch AI
- FrontierMath (Tiers 1-3)
- Frontiermath tier 4 — Epoch AI
- Gbaeval — Epoch AI
- Humanity's Last Exam (Epoch AI replication)
- Lambada — Epoch AI
- Lech mazur writing — Epoch AI
- Math level 5 — Epoch AI
- Metr time horizons — Epoch AI
- Mystery game puzzles — Epoch AI
- Piqa — Epoch AI
- Proofbench — Epoch AI
- Scicode — Epoch AI
- Science qa — Epoch AI
- Simplebench — Epoch AI
- Simpleqa verified — Epoch AI
- Surface evolver bench — Epoch AI
- Terminalbench — Epoch AI
- The agent company — Epoch AI
- Vending bench 2 — Epoch AI
- Webdev arena — Epoch AI
- Weirdml — Epoch AI
- Wino grande — Epoch AI
- GPQA Diamond
- LMArena · Agent
- LMArena · Search
- LMArena · WebDev
Capability is a model's mean percentile across the boards below, on each board's widest metric. Prices are quoted as you are billed them; where a model charges separately to read and to write, the two are ranked against each other at one part in to three parts out, which for this field puts the middle at $0.48 per million tokens. Best value for money is the most capable at or under that middle, and cheapest is the least expensive of those above the middle for capability — without that second floor the cheapest answer is reliably the worst model in the category. Figures read 2026-10-31.