Cost efficiency 1.1

Models: 21
Guesser cost range: 176×
Pareto-efficient: 8 of 21
Most efficient: GPT-6.1 Sol (high)

Official efficiency ranking

Distance from the lower-left ideal.

Question score and Guesser cost are each scaled from 0 (best among these models) to 1 (worst). The ranking weights both equally and measures the distance from the best possible point. Lower is better.

Distance on the shared scale · lower is betterSelect a model row to view its results

1. GPT-6.1 Sol (high): 0.165. 0.101 scaled question score · 0.131 scaled cost. View results for GPT-6.1 Sol (high)
2. Gemini 3.7 Flash (high): 0.263. 0.187 scaled question score · 0.185 scaled cost. View results for Gemini 3.7 Flash (high)
3. Claude Opus 5.5 (high): 0.280. 0.086 scaled question score · 0.267 scaled cost. View results for Claude Opus 5.5 (high)
4. Muse Spark 1.3 Contributor (high): 0.303. 0.303 scaled question score · 0.012 scaled cost. View results for Muse Spark 1.3 Contributor (high)
5. Claude Sonnet 5.5 (high): 0.313. 0.269 scaled question score · 0.161 scaled cost. View results for Claude Sonnet 5.5 (high)
6. GPT-6 Sol (high): 0.314. 0.282 scaled question score · 0.139 scaled cost. View results for GPT-6 Sol (high)
7. Gemini 3.8 Flash (high): 0.321. 0.077 scaled question score · 0.312 scaled cost. View results for Gemini 3.8 Flash (high)
8. GPT-5.6 Sol (high): 0.345. 0.284 scaled question score · 0.197 scaled cost. View results for GPT-5.6 Sol (high)
9. Claude Haiku 5.5 (high): 0.409. 0.409 scaled question score · 0.011 scaled cost. View results for Claude Haiku 5.5 (high)
10. Claude Opus 5 (high): 0.438. 0.284 scaled question score · 0.333 scaled cost. View results for Claude Opus 5 (high)
11. Gemini 4 Argon (high): 0.471. 0.000 scaled question score · 0.471 scaled cost. View results for Gemini 4 Argon (high)
12. GPT-5.6 Luna (high): 0.476. 0.475 scaled question score · 0.017 scaled cost. View results for GPT-5.6 Luna (high)
13. GPT-6 Luna (high): 0.484. 0.484 scaled question score · 0.006 scaled cost. View results for GPT-6 Luna (high)
14. Grok 4.6 (high): 0.539. 0.224 scaled question score · 0.490 scaled cost. View results for Grok 4.6 (high)
15. Grok 4.7 (high): 0.573. 0.553 scaled question score · 0.151 scaled cost. View results for Grok 4.7 (high)
16. gpt-oss-120B (high): 0.603. 0.553 scaled question score · 0.242 scaled cost. View results for gpt-oss-120B (high)
17. GLM-5.3-Flash (high): 0.619. 0.619 scaled question score · 0.000 scaled cost. View results for GLM-5.3-Flash (high)
18. GPT-6 Astra (high): 0.631. 0.000 scaled question score · 0.631 scaled cost. View results for GPT-6 Astra (high)
19. Muse Wonder (high): 0.693. 0.385 scaled question score · 0.577 scaled cost. View results for Muse Wonder (high)
20. MiniMax M3 (high): 1.000. 1.000 scaled question score · 0.031 scaled cost. View results for MiniMax M3 (high)
21. Claude Fable 5.1 (high): 1.004. 0.092 scaled question score · 1.000 scaled cost. View results for Claude Fable 5.1 (high)

Trade-off map

Cost and question score on a shared scale.

Both axes use the same 0-to-1 scale. Dashed curves connect points at equal distance from the lower-left corner.

Diamond Pareto-efficient: no other model is both cheaper and better Circle Other ranked model

Lower-left is betterCurves show equal ideal distance. Numbers match the table rank.

1. GPT-6.1 Sol (high)0.165 distance · Pareto-efficient
2. Gemini 3.7 Flash (high)0.263 distance
3. Claude Opus 5.5 (high)0.280 distance · Pareto-efficient
4. Muse Spark 1.3 Contributor (high)0.303 distance · Pareto-efficient
5. Claude Sonnet 5.5 (high)0.313 distance
6. GPT-6 Sol (high)0.314 distance
7. Gemini 3.8 Flash (high)0.321 distance · Pareto-efficient
8. GPT-5.6 Sol (high)0.345 distance
9. Claude Haiku 5.5 (high)0.409 distance · Pareto-efficient
10. Claude Opus 5 (high)0.438 distance
11. Gemini 4 Argon (high)0.471 distance · Pareto-efficient
12. GPT-5.6 Luna (high)0.476 distance
13. GPT-6 Luna (high)0.484 distance · Pareto-efficient
14. Grok 4.6 (high)0.539 distance
15. Grok 4.7 (high)0.573 distance
16. gpt-oss-120B (high)0.603 distance
17. GLM-5.3-Flash (high)0.619 distance · Pareto-efficient
18. GPT-6 Astra (high)0.631 distance
19. Muse Wonder (high)0.693 distance
20. MiniMax M3 (high)1.000 distance
21. Claude Fable 5.1 (high)1.004 distance

GPT-6.1 Sol (high): 15.23 questions, $0.0376 Guesser cost per round, ideal distance 0.165, rank 1. Pareto-efficient. View results for GPT-6.1 Sol (high)
Gemini 3.7 Flash (high): 16.57 questions, $0.0525 Guesser cost per round, ideal distance 0.263, rank 2. View results for Gemini 3.7 Flash (high)
Claude Opus 5.5 (high): 15.00 questions, $0.0751 Guesser cost per round, ideal distance 0.280, rank 3. Pareto-efficient. View results for Claude Opus 5.5 (high)
Muse Spark 1.3 Contributor (high): 18.37 questions, $0.0049 Guesser cost per round, ideal distance 0.303, rank 4. Pareto-efficient. View results for Muse Spark 1.3 Contributor (high)
Claude Sonnet 5.5 (high): 17.83 questions, $0.0460 Guesser cost per round, ideal distance 0.313, rank 5. View results for Claude Sonnet 5.5 (high)
GPT-6 Sol (high): 18.03 questions, $0.0398 Guesser cost per round, ideal distance 0.314, rank 6. View results for GPT-6 Sol (high)
Gemini 3.8 Flash (high): 14.87 questions, $0.0875 Guesser cost per round, ideal distance 0.321, rank 7. Pareto-efficient. View results for Gemini 3.8 Flash (high)
GPT-5.6 Sol (high): 18.07 questions, $0.0557 Guesser cost per round, ideal distance 0.345, rank 8. View results for GPT-5.6 Sol (high)
Claude Haiku 5.5 (high): 20.00 questions, $0.0046 Guesser cost per round, ideal distance 0.409, rank 9. Pareto-efficient. View results for Claude Haiku 5.5 (high)
Claude Opus 5 (high): 18.07 questions, $0.0934 Guesser cost per round, ideal distance 0.438, rank 10. View results for Claude Opus 5 (high)
Gemini 4 Argon (high): 13.67 questions, $0.1314 Guesser cost per round, ideal distance 0.471, rank 11. Pareto-efficient. View results for Gemini 4 Argon (high)
GPT-5.6 Luna (high): 21.03 questions, $0.0063 Guesser cost per round, ideal distance 0.476, rank 12. View results for GPT-5.6 Luna (high)
GPT-6 Luna (high): 21.17 questions, $0.0031 Guesser cost per round, ideal distance 0.484, rank 13. Pareto-efficient. View results for GPT-6 Luna (high)
Grok 4.6 (high): 17.13 questions, $0.1367 Guesser cost per round, ideal distance 0.539, rank 14. View results for Grok 4.6 (high)
Grok 4.7 (high): 22.23 questions, $0.0432 Guesser cost per round, ideal distance 0.573, rank 15. View results for Grok 4.7 (high)
gpt-oss-120B (high): 22.23 questions, $0.0682 Guesser cost per round, ideal distance 0.603, rank 16. View results for gpt-oss-120B (high)
GLM-5.3-Flash (high): 23.27 questions, $0.0016 Guesser cost per round, ideal distance 0.619, rank 17. Pareto-efficient. View results for GLM-5.3-Flash (high)
GPT-6 Astra (high): 13.67 questions, $0.1756 Guesser cost per round, ideal distance 0.631, rank 18. View results for GPT-6 Astra (high)
Muse Wonder (high): 19.63 questions, $0.1605 Guesser cost per round, ideal distance 0.693, rank 19. View results for Muse Wonder (high)
MiniMax M3 (high): 29.17 questions, $0.0100 Guesser cost per round, ideal distance 1.000, rank 20. View results for MiniMax M3 (high)
Claude Fable 5.1 (high): 15.10 questions, $0.2772 Guesser cost per round, ideal distance 1.004, rank 21. View results for Claude Fable 5.1 (high)

Efficiency ranking
Ideal-distance rank			Paretoefficient
Rank 1	GPT-6.1 Sol (high)openai	0.165	Yes	6	15.23	$0.0376	97%
Rank 2	Gemini 3.7 Flash (high)google-ai-studio	0.263		7	16.57	$0.0525	93%
Rank 3	Claude Opus 5.5 (high)anthropic	0.280	Yes	4	15.00	$0.0751	97%
Rank 4	Muse Spark 1.3 Contributor (high)meta	0.303	Yes	13	18.37	$0.0049	97%
Rank 5	Claude Sonnet 5.5 (high)anthropic	0.313		9	17.83	$0.0460	90%
Rank 6	GPT-6 Sol (high)openai	0.314		10	18.03	$0.0398	100%
Rank 7	Gemini 3.8 Flash (high)google-ai-studio	0.321	Yes	3	14.87	$0.0875	100%
Rank 8	GPT-5.6 Sol (high)openai	0.345		12	18.07	$0.0557	97%
Rank 9	Claude Haiku 5.5 (high)anthropic	0.409	Yes	15	20.00	$0.0046	83%
Rank 10	Claude Opus 5 (high)anthropic	0.438		11	18.07	$0.0934	93%
Rank 11	Gemini 4 Argon (high)google-vertex	0.471	Yes	2	13.67	$0.1314	100%
Rank 12	GPT-5.6 Luna (high)openai	0.476		16	21.03	$0.0063	90%
Rank 13	GPT-6 Luna (high)openai	0.484	Yes	17	21.17	$0.0031	80%
Rank 14	Grok 4.6 (high)xai	0.539		8	17.13	$0.1367	100%
Rank 15	Grok 4.7 (high)xai	0.573		18	22.23	$0.0432	73%
Rank 16	gpt-oss-120B (high)cerebras	0.603		19	22.23	$0.0682	67%
Rank 17	GLM-5.3-Flash (high)z-ai	0.619	Yes	20	23.27	$0.0016	73%
Rank 18	GPT-6 Astra (high)openai	0.631		1	13.67	$0.1756	100%
Rank 19	Muse Wonder (high)meta	0.693		14	19.63	$0.1605	97%
Rank 20	MiniMax M3 (high)coreweave	1.000		21	29.17	$0.0100	50%
Rank 21	Claude Fable 5.1 (high)anthropic	1.004		5	15.10	$0.2772	93%

GPT-6.1 Sol (high)openai0.165Ideal distance

Pareto-efficient: Yes
Question score: 15.23
Guesser cost / round: $0.0376

Gemini 3.7 Flash (high)google-ai-studio0.263Ideal distance

Pareto-efficient: No
Question score: 16.57
Guesser cost / round: $0.0525

Claude Opus 5.5 (high)anthropic0.280Ideal distance

Pareto-efficient: Yes
Question score: 15.00
Guesser cost / round: $0.0751

Muse Spark 1.3 Contributor (high)meta0.303Ideal distance

Pareto-efficient: Yes
Question score: 18.37
Guesser cost / round: $0.0049

Claude Sonnet 5.5 (high)anthropic0.313Ideal distance

Pareto-efficient: No
Question score: 17.83
Guesser cost / round: $0.0460

GPT-6 Sol (high)openai0.314Ideal distance

Pareto-efficient: No
Question score: 18.03
Guesser cost / round: $0.0398

Gemini 3.8 Flash (high)google-ai-studio0.321Ideal distance

Pareto-efficient: Yes
Question score: 14.87
Guesser cost / round: $0.0875

GPT-5.6 Sol (high)openai0.345Ideal distance

Pareto-efficient: No
Question score: 18.07
Guesser cost / round: $0.0557

Claude Haiku 5.5 (high)anthropic0.409Ideal distance

Pareto-efficient: Yes
Question score: 20.00
Guesser cost / round: $0.0046

Claude Opus 5 (high)anthropic0.438Ideal distance

Pareto-efficient: No
Question score: 18.07
Guesser cost / round: $0.0934

Gemini 4 Argon (high)google-vertex0.471Ideal distance

Pareto-efficient: Yes
Question score: 13.67
Guesser cost / round: $0.1314

GPT-5.6 Luna (high)openai0.476Ideal distance

Pareto-efficient: No
Question score: 21.03
Guesser cost / round: $0.0063

GPT-6 Luna (high)openai0.484Ideal distance

Pareto-efficient: Yes
Question score: 21.17
Guesser cost / round: $0.0031

Grok 4.6 (high)xai0.539Ideal distance

Pareto-efficient: No
Question score: 17.13
Guesser cost / round: $0.1367

Grok 4.7 (high)xai0.573Ideal distance

Pareto-efficient: No
Question score: 22.23
Guesser cost / round: $0.0432

gpt-oss-120B (high)cerebras0.603Ideal distance

Pareto-efficient: No
Question score: 22.23
Guesser cost / round: $0.0682

GLM-5.3-Flash (high)z-ai0.619Ideal distance

Pareto-efficient: Yes
Question score: 23.27
Guesser cost / round: $0.0016

GPT-6 Astra (high)openai0.631Ideal distance

Pareto-efficient: No
Question score: 13.67
Guesser cost / round: $0.1756

Muse Wonder (high)meta0.693Ideal distance

Pareto-efficient: No
Question score: 19.63
Guesser cost / round: $0.1605

MiniMax M3 (high)coreweave1.000Ideal distance

Pareto-efficient: No
Question score: 29.17
Guesser cost / round: $0.0100

Claude Fable 5.1 (high)anthropic1.004Ideal distance

Pareto-efficient: No
Question score: 15.10
Guesser cost / round: $0.2772

How it is calculated

Distance from the lowest score and cost.

Both measures use a 0-to-1 scale and have equal weight. Lower is better.

Formula√(scaled question score² + scaled Guesser cost²)

Calculation detailsSteps and limits

Scale question score as (value − lowest) ÷ (highest − lowest) across the compared models.
Scale Guesser cost per round with the same calculation.
Measure straight-line distance from (0, 0). Lower is better.

A model with scaled question score 0.06 and scaled cost 0.08 has distance √(0.06² + 0.08²) = 0.10.

The question score includes the fixed score for failed rounds.

Adding or removing a model can change every scaled value and rank.