llmcloud.ai
Workloads

We'll tell you which model to use.

A catalog is not an answer. For each workload we name one default, one cheap alternate and one escape hatch — plus the routing directive that encodes the same decision so you never have to hardcode a model.

WorkloadDefaultCheap alternateEscape hatchDirectiveEst. $/1M
Coding agentanthropic/claude-sonnet-4.5alibaba/qwen3-coder-480banthropic/claude-opus-4.5auto:coding$3.00 blended
Deep reasoning & mathopenai/gpt-5.5deepseek/deepseek-r1anthropic/claude-opus-4.5auto:reasoning$12.50 blended
Function calling & agentsopenai/gpt-5google/gemini-3.1-flashanthropic/claude-sonnet-4.5auto:tools$1.40 blended
Long-context retrievalgoogle/gemini-3.1-progoogle/gemini-3.1-flashopenai/gpt-5.5auto:long-context$7.00 blended
Vision & document parsinggoogle/gemini-3.1-proalibaba/qwen3-vl-235bopenai/gpt-5.5auto:vision$7.00 blended
Batch classification & extractionopenai/gpt-5.5-miniopenai/gpt-5.5-nanoanthropic/claude-haiku-4.5auto:cost$0.10–0.40
Realtime & voice turnsgroq/llama-4-70bgoogle/gemini-3.1-flashopenai/gpt-5.5-miniauto:speed$0.40 blended
High-volume product chatgoogle/gemini-3.1-flashdeepseek/deepseek-v4anthropic/claude-sonnet-4.5auto:cost + cache$0.28–0.55

The reasoning behind each pick

Coding agent

auto:coding

20–200k context, many tool calls, long diffs, latency tolerant

Highest diff fidelity per dollar; holds multi-file edits without re-reading the repo.

Watch out · Opus is 5x the price for ~3 points of quality — reserve it for failed retries.

→ Full coding agent leaderboard

Deep reasoning & math

auto:reasoning

Short prompt, long thinking budget, single answer, seconds to minutes

Reasoning tokens dominate the bill, so quality per thinking-token beats headline price.

Watch out · Cap the thinking budget — an uncapped reasoning run can cost 20x a normal call.

→ Full deep reasoning & math leaderboard

Function calling & agents

auto:tools

Many small turns, strict JSON, 5–40 tools in scope

Schema conformance under a wide tool set matters more than raw reasoning score.

Watch out · Provider variance is large here — the same model conforms differently per host.

→ Full function calling & agents leaderboard

Long-context retrieval

auto:long-context

200k–2M input, small output, recall-critical

Effective recall at depth, not the advertised window, is the deciding number.

Watch out · Above ~400k, chunk-and-rerank is usually cheaper and more accurate than one big call.

→ Full long-context retrieval leaderboard

Vision & document parsing

auto:vision

Images, PDFs, charts and screenshots in, structured data out

Chart and table extraction quality separates models far more than natural-image captioning.

Watch out · Image tokens are billed at very different rates per provider — check the receipt.

→ Full vision & document parsing leaderboard

Batch classification & extraction

auto:cost

Millions of short calls, fixed schema, throughput-bound

At this volume a 3x price difference dwarfs a 2-point quality difference.

Watch out · Cache the system prompt — it is often 80% of the tokens in a batch job.

Realtime & voice turns

auto:speed

Sub-300ms first token, short turns, interruption-tolerant

TTFT is the whole experience; specialty silicon beats a smarter model that starts late.

Watch out · Specialty hosts have thin capacity — always pin a second upstream for failover.

High-volume product chat

auto:cost + cache

Long-lived sessions, repeated system prompts, cost-sensitive

Semantic caching does more for the bill here than any model swap.

Watch out · Measure cache hit rate before optimising the model — 40% hits beats a cheaper model.