Long-context agent workflows

Scores as of 2026-10-19 (8 days old).

Depot Pick

Google AI (Gemini) · 8.43/10 for this job (curated estimate, high confidence)

Capability includes context window + needle quality. Gemini often wins value at 1M; Claude strong at high quality long context.

Runner-up

Anthropic · 8.42/10

Scores higher than the pick on: Capability fit, Latency / reliability, Developer & agent DX, Privacy / data retention.

When not to use the pick

Quality on tool use matters more than raw context — then Anthropic; enterprise GCP already → Vertex; want Kimi K3 1M ctx at $3/$15 — moonshot.

Lock-in / migration

The pick has lower stored lock-in risk than runner-up Anthropic (6 vs 6.5). Self-host fallback listed for this job: OpenAI MRCR.

Sourced price (usd_per_million_input_tokens)

Pick's lowest sourced SKU: gemini-3.5-flash-lite $0.3 (usd_per_million_input_tokens) — source, as of 2026-10-21

Cheapest matching SKU: Anthropic Haiku 5.5 (prompts up to 100K tokens) $0.1 — source, as of 2026-10-27, re-checked 2026-10-27. Not the pick; not quality-adjusted.

Price history: first seen 2026-10-21 · last changed no change recorded · history JSON

Free option

• OpenAI MRCR (oss) — source, as of 2026-10-25

Failure modes to watch

• Context overflow: If the input exceeds the window, say so and retrieve a narrower slice. Do not answer from a truncated head as if it were the whole source.
• Instruction drift: Restate the stop condition in the trace. End the run when it is met. Review long traces for steps that do not serve the original task.
• Lost in the middle: Ask the agent to name the passage it used. Spot-check questions whose only support sits in the middle of the context, not only at the edges.
• Runaway spend: Set a budget and a tool-call cap per task. Alert when spend jumps. Stop the run at the cap instead of summarizing a partial trace as success.
• Silent truncation: Tools must return whether the payload was truncated and how long the original was. Fail the task if the answer depends on the missing span.

Reported outcomes (submitted reports only)

This job: loading… JSON · reports never change scores · seed/test records excluded

Depot Pick (Google AI (Gemini)) on this job: loading… JSON · reports never change scores · seed/test records excluded

Same-job, same-unit comparisons: Google AI (Gemini) vs Anthropic · Google AI (Gemini) vs OpenAI · Google AI (Gemini) vs Moonshot (Kimi)

Cite: Agent Provider Depot, Depot Pick for long-context-agents: google-ai, scores as of 2026-10-19. Agents: GET /api/decide?jtbd=long-context-agents