DeepSeek V3
General-purpose chat + coding model with strong tool-use reliability. Default workhorse for most coding-agent loops.
Browse every model qlaud routes to. Customer-facing prices include our 7% markup. Models with multiple hosts can be requested with :nitro for the fastest tool-verified host or :floor for the cheapest live host.
Showing all 5 models
General-purpose chat + coding model with strong tool-use reliability. Default workhorse for most coding-agent loops.
Reasoning-tuned variant with separate chain-of-thought stream. Translates into Anthropic thinking blocks; rendered natively in Claude Code.
Frontier coding model — 87.6% on SWE-bench Verified. Best agentic-loop reliability. Routed via Anthropic native API: cache_control markers, image content blocks, and thinking blocks all preserved verbatim — Claude Code customers see ~75% input-cost reduction from prompt cache without changing client code.
Mid-tier Anthropic model. Strong tool use + vision; the default Claude Code CLI uses this slug. Native-passthrough route preserves prompt cache + image + thinking blocks.
Backwards-compatible alias for the slug Claude Code CLI ships with as default. Routed to Anthropic Sonnet 4.6 via native passthrough — same Anthropic model, same prompt cache, same agentic UX.
All routed through Cloudflare AI Gateway. We add new providers as Cloudflare adds them to its supported list.
Want the full machine-readable catalog? It's public and edge-cached:
curl https://api.qlaud.ai/v1/catalog