Run #701

success · fetched 2026-06-22 00:00:13 · 9.80 MB raw HTML · 540 models

Open this run in comparison view · JSON results

Top quality: Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) (63.1 pts)

Cost/task vs quality?

43 models · 10 on the Pareto frontier · combined quality vs Cost per Task

1024385266$50$100$200$500$1k$2k$5k cost per task, log scale combined quality Claude Opus 4.7 (Adaptive Reasoning, Max Effort) · Cost/task: $3,738 · Time/task: - · Quality: 57.2GPT-5.4 (xhigh) · Cost/task: $2,261 · Time/task: - · Quality: 54.5GPT-5.5 (medium) · Cost/task: $969 · Time/task: - · Quality: 53.2Gemini 3.5 Flash (high) · Cost/task: $1,142 · Time/task: - · Quality: 52.6Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) · Cost/task: $3,356 · Time/task: - · Quality: 50.3Qwen3.7 Max · Cost/task: $1,159 · Time/task: - · Quality: 47.5MiniMax-M3 · Cost/task: $235 · Time/task: - · Quality: 46.1Gemini 3.1 Pro Preview · Cost/task: $860 · Time/task: - · Quality: 45.6GPT-5.5 (low) · Cost/task: $390 · Time/task: - · Quality: 44.9Kimi K2.7 Code · Cost/task: $530 · Time/task: - · Quality: 44.1Kimi K2.6 · Cost/task: $839 · Time/task: - · Quality: 43.0GPT-5.4 mini (xhigh) · Cost/task: $1,158 · Time/task: - · Quality: 42.1GLM-5.1 (Reasoning) · Cost/task: $674 · Time/task: - · Quality: 41.9GPT-5.4 nano (xhigh) · Cost/task: $289 · Time/task: - · Quality: 40.6Qwen3.6 Plus · Cost/task: $484 · Time/task: - · Quality: 40.5Qwen3.6 27B (Reasoning) · Cost/task: $668 · Time/task: - · Quality: 39.3GPT-5.5 (Non-reasoning) · Cost/task: $223 · Time/task: - · Quality: 39.2MiniMax-M2.7 · Cost/task: $144 · Time/task: - · Quality: 38.8Qwen3.7 Plus · Cost/task: $152 · Time/task: - · Quality: 38.5Nemotron 3 Ultra 550B A55B (Reasoning) · Cost/task: $444 · Time/task: - · Quality: 38.1Grok 4.3 (high) · Cost/task: $319 · Time/task: - · Quality: 34.6Qwen3.5 397B A17B (Reasoning) · Cost/task: $528 · Time/task: - · Quality: 33.9Qwen3.5 122B A10B (Reasoning) · Cost/task: $447 · Time/task: - · Quality: 32.9Mistral Medium 3.5 · Cost/task: $1,014 · Time/task: - · Quality: 31.9Qwen3.6 35B A3B (Reasoning) · Cost/task: $333 · Time/task: - · Quality: 31.6Ring-2.6-1T · Cost/task: $459 · Time/task: - · Quality: 30.8Claude 4.5 Haiku (Reasoning) · Cost/task: $539 · Time/task: - · Quality: 30.0Step 3.7 Flash · Cost/task: $320 · Time/task: - · Quality: 29.5Grok 4.3 (Non-reasoning) · Cost/task: $297 · Time/task: - · Quality: 27.6NVIDIA Nemotron 3 Super 120B A12B (Reasoning) · Cost/task: $287 · Time/task: - · Quality: 23.9gpt-oss-120b (high) · Cost/task: $96.3 · Time/task: - · Quality: 22.5Gemini 3.1 Flash-Lite · Cost/task: $94.8 · Time/task: - · Quality: 22.0Nova 2.0 Pro Preview (medium) · Cost/task: $407 · Time/task: - · Quality: 20.9Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) · Cost/task: $6,228 · Time/task: - · Quality: 63.1 · Pareto frontierClaude Opus 4.8 (Adaptive Reasoning, Max Effort) · Cost/task: $4,012 · Time/task: - · Quality: 59.0 · Pareto frontierGPT-5.5 (xhigh) · Cost/task: $2,588 · Time/task: - · Quality: 58.2 · Pareto frontierGPT-5.5 (high) · Cost/task: $1,775 · Time/task: - · Quality: 56.1 · Pareto frontierGLM-5.2 (max) · Cost/task: $869 · Time/task: - · Quality: 54.3 · Pareto frontierDeepSeek V4 Pro (Reasoning, Max Effort) · Cost/task: $180 · Time/task: - · Quality: 46.7 · Pareto frontierMiMo-V2.5-Pro · Cost/task: $99.1 · Time/task: - · Quality: 43.8 · Pareto frontierDeepSeek V4 Flash (Reasoning, Max Effort) · Cost/task: $78.4 · Time/task: - · Quality: 42.5 · Pareto frontierGemma 4 26B A4B (Reasoning) · Cost/task: $54.5 · Time/task: - · Quality: 25.3 · Pareto frontiergpt-oss-20B (high) · Cost/task: $29.9 · Time/task: - · Quality: 12.9 · Pareto frontier gpt-oss-20B (high)Gemma 4 26B A4BDeepSeek V4 Flash…MiMo-V2.5-ProDeepSeek V4 Pro (…GLM-5.2 (max)GPT-5.5 (high)GPT-5.5 (xhigh)Claude Opus 4.8 (…Claude Fable 5 (w…
#?Pareto?Model?Released?Cost/task?Time/task?$/Q?Qual?ΔTop?Intel?Code?Agent?Pen?Score?
1 Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)
Anthropic
2026-06-09 $6,228 - 98.7659 63.1 0.0 59.9 76.5 52.8 57.9 63.1
2 Claude Opus 4.8 (Adaptive Reasoning, Max Effort)
Anthropic
2026-05-28 $4,012 - 67.9464 59.0 -4.0 55.7 74.3 47.2 56.0 59.0
3 GPT-5.5 (xhigh)
OpenAI
2026-04-23 $2,588 - 44.4736 58.2 -4.9 54.8 74.9 44.9 54.1 58.2
4 Claude Opus 4.7 (Adaptive Reasoning, Max Effort)
Anthropic
2026-04-16 $3,738 - 65.3777 57.2 -5.9 53.5 73.6 44.4 55.7 57.2
5 GPT-5.5 (high)
OpenAI
2026-04-23 $1,775 - 31.6742 56.1 -7.0 53.1 71.6 43.5 52.5 56.1
6 GPT-5.4 (xhigh)
OpenAI
2026-03-05 $2,261 - 41.4844 54.5 -8.5 51.4 71.1 41.1 53.5 54.5
7 GLM-5.2 (max)
Z AI
2026-06-16 $869 - 16.0010 54.3 -8.8 51.1 68.8 43.1 49.4 54.3
8 GPT-5.5 (medium)
OpenAI
2026-04-23 $969 - 18.1967 53.2 -9.8 50.4 71.5 37.8 49.9 53.2
9 Gemini 3.5 Flash (high)
Google
2026-05-19 $1,142 - 21.7049 52.6 -10.5 50.2 70.1 37.4 50.6 52.6
10 Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)
Anthropic
2026-02-17 $3,356 - 66.6696 50.3 -12.7 47.2 63.0 40.8 55.3 50.3
11 Qwen3.7 Max
Alibaba
2026-05-19 $1,159 - 24.3854 47.5 -15.5 46.0 66.0 30.6 50.6 47.5
12 DeepSeek V4 Pro (Reasoning, Max Effort)
DeepSeek
2026-04-24 $180 - 3.8530 46.7 -16.4 44.3 59.4 36.4 42.5 46.7
13 MiniMax-M3
MiniMax
2026-06-01 $235 - 5.1000 46.1 -16.9 44.4 58.6 35.4 43.7 46.1
14 Gemini 3.1 Pro Preview
Google
2026-02-19 $860 - 18.8718 45.6 -17.5 46.5 68.8 21.4 49.3 45.6
15 GPT-5.5 (low)
OpenAI
2026-04-23 $390 - 8.6765 44.9 -18.1 43.5 60.9 30.4 45.9 44.9
16 Kimi K2.7 Code
Kimi
2026-06-12 $530 - 12.0260 44.1 -19.0 41.9 60.8 29.6 47.2 44.1
17 MiMo-V2.5-Pro
Xiaomi
2026-04-22 $99.1 - 2.2600 43.8 -19.2 42.2 60.2 29.1 40.0 43.8
18 Kimi K2.6
Kimi
2026-04-20 $839 - 19.4813 43.0 -20.0 42.8 56.0 30.3 49.2 43.0
19 DeepSeek V4 Flash (Reasoning, Max Effort)
DeepSeek
2026-04-24 $78.4 - 1.8450 42.5 -20.6 40.3 56.2 31.1 38.9 42.5
20 GPT-5.4 mini (xhigh)
OpenAI
2026-03-17 $1,158 - 27.5117 42.1 -21.0 40.0 56.1 30.2 50.6 42.1
21 GLM-5.1 (Reasoning)
Z AI
2026-04-07 $674 - 16.0715 41.9 -21.1 40.2 55.8 29.9 48.3 41.9
22 GPT-5.4 nano (xhigh)
OpenAI
2026-03-17 $289 - 7.1068 40.6 -22.4 38.2 56.1 27.5 44.6 40.6
23 Qwen3.6 Plus
Alibaba
2026-04-02 $484 - 11.9455 40.5 -22.5 39.6 54.5 27.6 46.9 40.5
24 Qwen3.6 27B (Reasoning)
Alibaba
2026-04-22 $668 - 17.0141 39.3 -23.8 37.1 53.7 27.0 48.2 39.3
25 GPT-5.5 (Non-reasoning)
OpenAI
2026-04-23 $223 - 5.6872 39.2 -23.8 35.4 56.5 25.8 43.5 39.2
26 MiniMax-M2.7
MiniMax
2026-03-18 $144 - 3.7067 38.8 -24.3 38.1 52.6 25.6 41.6 38.8
27 Qwen3.7 Plus
Alibaba
2026-06-01 $152 - 3.9469 38.5 -24.5 39.0 55.9 20.8 41.8 38.5
28 Nemotron 3 Ultra 550B A55B (Reasoning)
NVIDIA
2026-06-04 $444 - 11.6348 38.1 -24.9 37.8 49.3 27.4 46.5 38.1
29 Grok 4.3 (high)
xAI
2026-04-30 $319 - 9.1956 34.6 -28.4 37.6 42.2 24.1 45.0 34.6
30 Qwen3.5 397B A17B (Reasoning)
Alibaba
2026-02-16 $528 - 15.5709 33.9 -29.1 33.7 48.2 19.8 47.2 33.9
31 Qwen3.5 122B A10B (Reasoning)
Alibaba
2026-02-24 $447 - 13.5872 32.9 -30.2 32.3 45.7 20.7 46.5 32.9
32 Mistral Medium 3.5
Mistral
2026-04-29 $1,014 - 31.7482 31.9 -31.1 29.9 46.9 19.0 50.1 31.9
33 Qwen3.6 35B A3B (Reasoning)
Alibaba
2026-04-16 $333 - 10.5209 31.6 -31.4 31.6 41.9 21.4 45.2 31.6
34 Ring-2.6-1T
InclusionAI
2026-05-08 $459 - 14.9282 30.8 -32.3 30.6 42.8 18.9 46.6 30.8
35 Claude 4.5 Haiku (Reasoning)
Anthropic
2025-10-15 $539 - 17.9938 30.0 -33.1 29.6 43.9 16.4 47.3 30.0
36 Step 3.7 Flash
StepFun
2026-05-29 $320 - 10.8487 29.5 -33.5 29.7 37.3 21.5 45.1 29.5
37 Grok 4.3 (Non-reasoning)
xAI
2026-04-30 $297 - 10.7820 27.6 -35.5 24.8 35.2 22.8 44.7 27.6
38 Gemma 4 26B A4B (Reasoning)
Google
2026-04-02 $54.5 - 2.1528 25.3 -37.7 25.7 39.3 11.0 37.4 25.3
39 NVIDIA Nemotron 3 Super 120B A12B (Reasoning)
NVIDIA
2026-03-11 $287 - 11.9745 23.9 -39.1 25.4 37.7 8.7 44.6 23.9
40 gpt-oss-120b (high)
OpenAI
2025-08-05 $96.3 - 4.2824 22.5 -40.6 23.8 30.4 13.2 39.8 22.5
41 Gemini 3.1 Flash-Lite
Google
2026-03-03 $94.8 - 4.3165 22.0 -41.1 25.0 34.7 6.2 39.8 22.0
42 Nova 2.0 Pro Preview (medium)
Amazon
2025-11-27 $407 - 19.4555 20.9 -42.1 21.8 34.0 7.0 46.1 20.9
43 gpt-oss-20B (high)
OpenAI
2025-08-05 $29.9 - 2.3165 12.9 -50.2 14.9 20.7 3.1 34.8 12.9