Which LLM gives the most benchmark quality per dollar?

Laguna S 2.1, at 561.60 quality points per dollar per 1M tokens, then DeepSeek V4.1 Flash and Nex-N2-Pro. Price is the 3:1 input-to-output tokens blend, not the headline input rate.

Why does this list only include models scoring 70 or more?

Quality is bounded at 100 while price runs to nearly zero, so an unfiltered quality-per-dollar ratio ranks the cheapest weak model first every time. The 70-point floor is what makes this a value list rather than a price list wearing a benchmark's clothes. 23 models clear it.

What is the cheapest LLM that still scores 70+?

Laguna S 2.1 at $0.125 per 1M tokens blended, scoring 70.2. Sorting this same pool on price rather than on quality per dollar barely changes it: no model moves more than two places. For the whole catalogue on raw price with no quality floor, see the cheapest models page.

The Best Value LLMs per Dollar

Which model buys the most benchmark quality per dollar?

23 models clear the 70-point floor and carry a metered price. The floor is an editorial line: without it the ratio always crowns the cheapest weak model. The cheapest of them outright is Laguna S 2.1 at $0.125 per 1M tokens blended.

Method: Quality score divided by the blended price per 1M tokens (3:1 input-to-output tokens: 0.75 x input + 0.25 x output), among models scoring at least 70. Models priced at $0 (open weights with no metered API) are left out: you cannot divide by a price that does not exist. Pricing as of October 2026. 4 rows below carry a score measured at a non-default reasoning tier and are marked as such; every affected model is named on the hub. Read the full method.

Rank	Model	Quality per $/1M	Quality	Blended $/1M	Evidence
#1	Laguna S 2.1Poolside	561.60	70.2	$0.125	Terminal-Bench only
#2	DeepSeek V4.1 FlashDeepSeek	345.14	90.6	$0.263	Terminal-Bench only
#3	Nex-N2-ProNex AGI	75.60	75.6	$1.00	Both suites
#4	Nemotron 3 Ultra 550BNVIDIA	71.90	71.9	$1.00	SWE-bench only
#5	MiMo-V2-ProXiaomi	52.00	78.0	$1.50	SWE-bench only
#6	Gemini 3.6 FlashGoogle	51.67	77.5	$1.50	Terminal-Bench only
#7	Muse Spark 1.3Meta	44.40	88.8	$2.00	Terminal-Bench only
#8	Muse Spark 1.1Meta	38.95	77.9	$2.00	Terminal-Bench only
#9	GLM-5.2Zhipu· non-default tier	36.23	77.9	$2.15	Terminal-Bench only
#10	Fugu MaxSakana AI	29.83	89.5	$3.00	Terminal-Bench only
#11	Grok 4.5xAI	27.20	81.6	$3.00	Terminal-Bench only
#12	Gemini 3.5 FlashGoogle	22.58	76.2	$3.38	Terminal-Bench only
#13	Claude Sonnet 5Anthropic· non-default tier	20.83	83.3	$4.00	Both suites
#14	Qwen3.7 MaxAlibaba	19.87	74.5	$3.75	Terminal-Bench only
#15	GPT-5.6 TerraOpenAI	16.07	72.3	$4.50	Terminal-Bench only
#16	Kimi K3Moonshot	14.17	85.0	$6.00	Terminal-Bench only
#17	Claude Sonnet 4.6Anthropic· non-default tier	11.87	71.2	$6.00	Terminal-Bench only
#18	GPT-5.6 SolOpenAI	10.76	86.1	$8.00	Terminal-Bench only
#19	Claude Opus 5Anthropic	9.20	92.0	$10.00	Both suites
#20	Claude Opus 4.8Anthropic	8.70	87.0	$10.00	Both suites
#21	Claude Opus 4.7Anthropic· non-default tier	8.31	83.1	$10.00	Terminal-Bench only
#22	GPT-5.5OpenAI	7.16	80.5	$11.25	Terminal-Bench only
#23	Claude Fable 5Anthropic	4.23	84.6	$20.00	Terminal-Bench only

Frequently asked

Nearby questions, answered elsewhere

Cheapest models on raw price

The whole catalogue on a plain 1:1 input/output average, with no quality floor at all. This page keeps the floor, so it never recommends something cheap that cannot do the work.

Highest score regardless of price

The quality ranking with no price term in it at all.

All rankings

The hub and the method The Best LLMs by Benchmark Score The Best LLMs by Context Window The Cheapest LLMs With a 1M-Token Context Window The Best LLMs on Terminal-Bench 2.1 The Best LLMs on Terminal-Bench Hard The Best LLMs on SWE-bench Verified The Best LLMs on GPQA Diamond The Best LLMs for Tool Use on tau2-bench