mario0369/llm-cost-same-prompt

Measured per-call LLM cost — same prompt, every model

Vendors publish prices per million tokens. Nobody publishes what one call actually costs, because that depends on how many tokens the model chooses to emit — and on the same question models differ by more than an order of magnitude. One model finishes a JSON extraction in 23 tokens; another writes 300.

This dataset sends a fixed set of prompts to every model at temperature 0, every night, and records the cost computed from the token usage each provider actually reported. Because the question is identical, the difference is the model, not the workload.

Current results (window: 30 days, as of 2026-10-08)

Median spread across tasks: 36.8x. Clean runs in window: 433.

Two-sentence explanation of a concept (short_answer)

Cheapest: deepseek-v4-flash. The most expensive costs 36.8x more for the same question.

model	USD per call	tokens in / out	latency (ms)	runs
deepseek-v4-flash	0.000129	45 / 83	3703	29
claude-sonnet-5	0.001361	195 / 97	7806	28
claude-opus-5	0.002552	45 / 93	7203	29
gpt-5.6-sol	0.003353	232 / 73	13695	14
grok-4.6	0.004035	824 / 398	14256	17
Kimi-K3	0.004752	108 / 295	14333	28

Compress a technical passage into three bullets (summarize)

Cheapest: deepseek-v4-flash. The most expensive costs 25.2x more for the same question.

model	USD per call	tokens in / out	latency (ms)	runs
deepseek-v4-flash	0.000285	194 / 151	4229	30
gpt-5.6-terra	0.002446	473 / 125	9001	1
claude-sonnet-5	0.002476	406 / 166	9235	29
claude-opus-5	0.005334	263 / 161	9722	28
Kimi-K3	0.005950	257 / 345	12099	27
grok-4.6	0.006428	836 / 793	17087	17
gpt-5.6-sol	0.007184	499 / 156	19334	12

Extract fields, return JSON only (structured_json)

Cheapest: deepseek-v4-flash. The most expensive costs 44.3x more for the same question.

model	USD per call	tokens in / out	latency (ms)	runs
deepseek-v4-flash	0.000078	73 / 35	3083	30
claude-sonnet-5	0.000719	225 / 27	6686	29
claude-opus-5	0.001100	84 / 27	3924	28
Kimi-K3	0.002050	137 / 109	10701	28
gpt-5.6-sol	0.002890	347 / 39	13880	15
grok-4.6	0.003456	564 / 388	7615	14

A routing artefact worth knowing about

Some requests arrive on a channel serving a large cached preamble: sending the single word hi comes back reporting ~4,400 prompt tokens, 3,840 of them cached. It is not stable — the same model and the same question sometimes hits it and sometimes does not, so it is a property of routing, not of the model. Those runs are excluded from the comparison above and reported here instead. They are still in probe_runs.csv (cached_prompt_tokens >= 1000), so you can check the exclusion rule yourself.

model	share of runs hitting it	USD when hit	USD when clean
gpt-5.6-terra	88%	0.017133	0.002446
gpt-5.6-sol	49%	0.024369	0.004305
claude-opus-5	3%	0.024190	0.002990
grok-4.6	45%	0.008369	0.004714

Method

• Identical prompt per task, temperature: 0, capped max_tokens. The prompts are in tasks.js and are deliberately timeless — they never change, so numbers stay comparable across months.
• Cost = tokens reported by the provider x the price actually paid. Responses without a usage field are discarded, not estimated — an estimate inside a "measured cost" dataset is a lie.
• One run per model per task per night; samples accumulate.
• probe_runs.csv contains every run, including failures and preamble hits, so the published averages can be recomputed from scratch.

What this does not tell you

Cost, not quality. Cheapest here says nothing about whether the answer is good. It also reflects these three prompts specifically — your prompts have a different input/output ratio, and that ratio is exactly what drives the difference. Use it as a starting point for bulk work, then measure your own.

Live version, updated nightly

• Human-readable, per-task pages: https://ainetcafe.com/costs
• JSON API, no key, CORS open: https://ainetcafe.com/api/model-costs
• Raw runs as CSV: https://ainetcafe.com/api/model-costs/raw.csv
• Callable from a bare URL by any agent: https://ainetcafe.com/t/model_costs?days=30

Citation

AI NetCafe real model cost dataset. https://ainetcafe.com/costs (CC BY 4.0)