Skip to content

AI models as each provider exposes them

Context window, maximum output, price and prompt caching for 6,186 entries. Built on models.dev, with 67 entries corrected or evaluated by providers.dev.

333 providers, 18 by subscription466 models86 cache evaluationsUpdated 2026-11-17

4 entries

This view as JSON/api/v1/entries?provider=atomic-chat&sort=verified&order=desc

Provider-model entries
Input window Cache lifetime
Qwen 3.5 9B (MLX 4-bit)
Qwen3_5-9B-MLX-4bit
Atomic Chat 32.77Kdocumented 32.77K 8.19K $0 / $0 2026-02-23 ShortUnknownLongUnknown models.dev2026-11-17
Qwen 3.5 9B (Q4_K_M)
Qwen3_5-9B-Q4_K_M
Atomic Chat 32.77Kdocumented 32.77K 8.19K $0 / $0 2026-02-23 ShortUnknownLongUnknown models.dev2026-11-17
Gemma 4 E4B Instruct (IQ4_XS)
gemma-4-E4B-it-IQ4_XS
Atomic Chat 32.77Kdocumented 32.77K 8.19K $0 / $0 2026-04-02 ShortUnknownLongUnknown models.dev2026-11-17
Gemma 4 E4B Instruct (MLX 4-bit)
gemma-4-E4B-it-MLX-4bit
Atomic Chat 32.77Kdocumented 32.77K 8.19K $0 / $0 2026-04-02 ShortUnknownLongUnknown models.dev2026-11-17