One field in the request made our agent 3x cheaper and 8x faster

Description and use

Specifying a reasoning effort level in the LLM request reduced Altair agent costs and execution time, improving task success rates.

Problem

Reasoning models can spend excessive time and tokens when the level of effort is not specified, causing high cost and slow performance.

Outcome

Specifying a low reasoning effort cut reasoning tokens from 9,743 to 235 per hard task (41×), cost from 12.6 m$ to 4.2 m$ (3×), and time from 4.1 min to 0.55 min (7.5×). Adaptive mode achieved the same success rate as always‑high with lower cost and faster time.

Workflow

• Run Altair on hard tasks with no reasoning level
• Measure cost, time, and success
• Add "reasoning":{"effort":"low"} to request
• Re‑run tasks and record metrics
• Test adaptive modes: always low, always high, high first step, low then high after failure
• Select best mode
• Update Altair code to set reasoning level per provider

Evidence

One field in the request body ("reasoning": {"effort": "low"}) made a task 3x cheaper and 7–8x faster

Source

Reasoning models decide by themselves how long to think if you don't tell them.

Source

12.6 m$ without a level, 4.2 m$ with "low": three times less.

Source

9,743 reasoning tokens per task on average without a level, 235 with "low" — 41x fewer.

Source

"think after a failure" solved as many as always‑high, at the cost and speed of low.

Source