PipelineScore
User Leaderboard

Every run. Every user.

480 submissions from 55 users. Sort, filter, find someone who ran your model.

Want on the board? Run the CLI.
Filter
Allalibabaanthropiccoheredatabricksdeepseekgooglemetamicrosoftmistralnousopenai|TRUNKMAINLINEFEEDERTAPDRIP|Lab-verified only
#User ↕Model ↕Score ↓Tier ↕
1reason-firstClaude Opus 4.792.6TRUNK
2devnullClaude Opus 4.792.5TRUNK
3labLabClaude Opus 4.792.3TRUNK
4shipfastlyClaude Opus 4.792.2TRUNK
5goose-engineerClaude Opus 4.792.1TRUNK
6tokenwranglerClaude Opus 4.791.9TRUNK
7silicon-djinnClaude Opus 4.791.6TRUNK
8dr_inferenceClaude Opus 4.790.9TRUNK
9devnullClaude Opus 4.790.5TRUNK
10temp-zeroGPT-5.590.4TRUNK
11qwen-stanClaude Opus 4.790.4TRUNK
12gemini-fanGPT-5.590.2TRUNK
13one-shot-onlyGPT-5.590.1TRUNK
14mira-shenClaude Opus 4.790.1TRUNK
15opus-believerClaude Opus 4.790.1TRUNK
16labLabGPT-5.590.1TRUNK
17deepseek-winsGPT-5.589.9MAINLINE
18labLabDeepSeek R1 671B-A37B89.5MAINLINE
19frontiermodelGPT-5.589.2MAINLINE
20context-windowGPT-5.589.0MAINLINE
21mira-shenGPT-5.588.9MAINLINE
22cpu-only-prophetQwen 3 235B-A22B MoE88.8MAINLINE
23opus-believerGPT-5.588.5MAINLINE
24dr_inferenceGPT-5.588.5MAINLINE
25goose-engineerGPT-5.588.4MAINLINE
26deepseek-winsGPT-5.588.2MAINLINE
27flash-attn-fanDeepSeek V3 671B-A37B88.1MAINLINE
28goose-engineerGemini 2.5 Pro87.5MAINLINE
29gpt-cynicGemini 2.5 Pro87.4MAINLINE
30labLabDeepSeek Coder V2 236B87.3MAINLINE
31edge_runnerGemini 2.5 Pro86.8MAINLINE
32rig-testerDeepSeek R1 671B-A37B86.6MAINLINE
33context-windowGemini 2.5 Pro86.5MAINLINE
34mira-shenGemini 2.5 Pro86.0MAINLINE
35mac-stackDeepSeek R1 671B-A37B86.0MAINLINE
36open-weightsGemini 2.5 Pro85.7MAINLINE
37silicon-djinnGemini 2.5 Pro85.7MAINLINE
38llama-llamaGemini 2.5 Pro85.4MAINLINE
39labLabGemini 2.5 Pro85.3MAINLINE
40quant-or-dieDeepSeek Coder V2 236B85.3MAINLINE
41gpt-cynicGemini 2.5 Pro85.3MAINLINE
42judgementalGemini 2.5 Pro85.2MAINLINE
43gpu-shopperQwen 2.5 72B Instruct85.1MAINLINE
44judgementalGemini 2.5 Pro85.0MAINLINE
45labLabQwen 3 72B Instruct85.0MAINLINE
46quant-or-dieHermes 3 Llama 3.1 405B84.9MAINLINE
47local-llama-fanHermes 3 Llama 3.1 405B84.8MAINLINE
48cpu-only-prophetQwen 2.5 72B Instruct84.8MAINLINE
49inference-monkDeepSeek Coder V2 236B84.7MAINLINE
50bench-ratDeepSeek Coder V2 236B84.5MAINLINE
51labLabQwen 3 235B-A22B MoE84.0MAINLINE
52deepseek-winsDeepSeek V484.0MAINLINE
53inference-monkDeepSeek R1 Distill Qwen 32B84.0MAINLINE
54sram-hoarderQwen 3 72B Instruct83.8MAINLINE
55labLabDeepSeek V483.7MAINLINE
56reason-firstDeepSeek V483.4MAINLINE
57rust_witchDeepSeek V483.3MAINLINE
58tensor-tomasLlama 3.3 70B Instruct83.3MAINLINE
59tensor-tomasDeepSeek V3 671B-A37B83.2MAINLINE
60labLabHermes 3 Llama 3.1 405B83.1MAINLINE
61judgementalDeepSeek V483.1MAINLINE
62local-llama-fanDeepSeek R1 671B-A37B83.0MAINLINE
63inference-monkQwen 3 235B-A22B MoE83.0MAINLINE
64qwen-stanDeepSeek V482.7MAINLINE
65cohere-cultDeepSeek V482.5MAINLINE
66labLabQwen 3 32B Instruct82.4MAINLINE
67apple-siliconQwen 3 235B-A22B MoE82.3MAINLINE
68rig-testerDeepSeek R1 Distill Qwen 32B82.3MAINLINE
69flash-attn-fanQwen 3 32B Instruct82.3MAINLINE
70edge_runnerDeepSeek V482.2MAINLINE
71labLabCommand A82.0MAINLINE
72quant-or-dieLlama 4 70B Instruct81.9MAINLINE
73open-weightsDeepSeek V481.9MAINLINE
74judgementalDeepSeek V481.9MAINLINE
75mac-stackQwen 2.5 Coder 32B81.8MAINLINE
76gpt-cynicDeepSeek V481.7MAINLINE
77qwen-stanQwen 3.6 72B81.6MAINLINE
78flash-attn-fanHermes 3 Llama 3.1 405B81.5MAINLINE
79open-weightsDeepSeek V481.5MAINLINE
80rocm-riderQwen 3 72B Instruct81.5MAINLINE
81gpt-cynicQwen 3.6 72B81.3MAINLINE
82tokenwranglerQwen 3.6 72B81.2MAINLINE
83labLabLlama 4 70B Instruct81.2MAINLINE
84labLabMixtral 8x22B Instruct81.2MAINLINE
85labLabDeepSeek R1 Distill Qwen 32B81.2MAINLINE
86cli-nativeQwen 3.6 72B81.1MAINLINE
87mac-stackQwen 2.5 32B Instruct81.0MAINLINE
88midnight-bencherLlama 4 70B Instruct81.0MAINLINE
89labLabDeepSeek V3 671B-A37B80.8MAINLINE
90cli-nativeQwen 3.6 72B80.8MAINLINE
91tps-counterQwen 3 72B Instruct80.8MAINLINE
92sram-hoarderQwen 2.5 72B Instruct80.6MAINLINE
93vram-monkDBRX Instruct 132B-MoE80.5MAINLINE
94judgementalQwen 3.6 72B80.5MAINLINE
95tensor-tomasCommand A80.3MAINLINE
96midnight-bencherQwen 2.5 VL 72B80.0MAINLINE
97cli-nativeMistral Large 280.0MAINLINE
98gguf-pilgrimQwen 2.5 Coder 32B80.0MAINLINE
99labLabWizardLM 2 8x22B80.0MAINLINE
100gpt-cynicMistral Large 280.0MAINLINE
1–100 of 480
Next →

Top reviewers

By best PipelineScore submitted. Click a user to open their dashboard.

#1
reason-first
4 runs
92.6
#2
devnull
4 runs
92.5
#3
lab
100 runs
92.3
#4
shipfastly
1 run
92.2
#5
goose-engineer
4 runs
92.1
#6
tokenwrangler
4 runs
91.9
#7
silicon-djinn
5 runs
91.6
#8
dr_inference
4 runs
90.9
#9
temp-zero
2 runs
90.4
#10
qwen-stan
4 runs
90.4
#11
gemini-fan
3 runs
90.2
#12
one-shot-only
2 runs
90.1
← Switch to model leaderboard