cheapos
COMPUTE OBSERVATORY · TELEMETRY MACRO-DECK

The cheapoS
Engine Room.

Usage shared by participating cheapoS installations: models, agent roles, and compute categories. These are reported token counts, not performance benchmarks.

Find free models, published benchmarks, and setup guidance →

Connect your models in cheapoS →
Total community compute250,750,936Reported tokens on the public board
Models shared63in the profiles summarized below
Roles reported4Worker, Reviewer, Planner, Coordinator
Public board members1Participating accounts on the public board

Input vs output

Input is context sent to models, including repeated context. Output is what models generate. Token volume measures usage, not work quality.

Input
246,815,36698.4%
Output
3,935,5701.6%

7,422 reported requests across 1 reporting public profile. Public-free, included-access and local usage only.

Cached input & reasoning output
Cached input
27,254,813 tokens · 1,355 of 7,422 requests reported
Reasoning output
580,329 tokens · 1,321 of 7,422 requests reported

These reported subsets are already included in input or output. Missing details stay unknown; they are not added to the total.

COGNITIVE DIVISION OF LABOR

Autonomous Workshop Roles

How autonomous cheapoS agent swarms divide their cognitive workloads when executing software tasks.

🧑‍💻73.9% share

Worker

The execution workhorse: editing files, running terminal checks, and tool calling.

Input
176,783,951
Output
1,692,469
178.48M178,476,420 tokens
🛡️21% share

Reviewer

The quality gatekeeper: auditing diffs, inspecting tests, and enforcing safety.

Input
49,216,215
Output
1,642,827
50.86M50,859,042 tokens
🗺️5.1% share

Planner

The architect: breaking complex requests into actionable autonomous milestones.

Input
11,860,762
Output
364,708
12.23M12,225,470 tokens
🧭0% share

Coordinator

The dispatcher: routing tasks between subagents and orchestrating checkpoints.

Input
107,308
Output
3,875
111.2K111,183 tokens
About these breakdowns: Role, model, and category totals below summarize all 2 members who share their usage. Model names appear only when members share them. Missing reports are left empty; token volume does not measure task quality.
THE MODEL POWER RANKINGS

Community Model Index

Model usage shared by members in the profiles summarized above. Rankings reflect token volume, not model quality or speed.

Provider details available for 57 of 63 model labels. Names below show recorded routes; older usage may lack provider details. Token totals include all shared history for each model, including when filtered by provider.

#Model NameRecorded providersCompute ShareTotal TokensAccess & CostReliability
1
gemini-3.1-flash-lite
antigravity via OmniRoute
19.9%
49.89M
Input
49,634,387
Output
251,700
Access not reported99.3%
2
dots-studio/dots-3-note-previewopenrouter/dots-studio/dots-3-note-preview:free
OpenRouter via OmniRoute
15.8%
39.56M
Input
39,288,603
Output
273,525
Public Free ($0.00)78.3%
3
groq/qwen/qwen3.6-27b
groq via OmniRoute
11.1%
27.74M
Input
27,009,398
Output
729,586
Public Free ($0.00)95.1%
4
antigravity/gemini-3.1-flash-lite
antigravity via OmniRoute
8.8%
22.12M
Input
21,905,162
Output
216,114
Access not reported89.2%
5
openai/gpt-oss-120b
groq via OmniRoute
8.4%
21.07M
Input
20,753,493
Output
314,495
Public Free ($0.00)100%
6
groq/qwen/qwen3.8-27b
groq via OmniRoute
4.4%
10.99M
Input
10,835,393
Output
151,638
Public Free ($0.00)93.7%
7
dots-studio/dots-3-note-previewdots-studio/dots-3-note-preview:free
OpenRouter via OmniRoute
3.6%
8.94M
Input
8,830,196
Output
105,288
Public Free ($0.00)100%
8
qwen/qwen3.8-27b
groq via OmniRoute
2.9%
7.24M
Input
7,101,403
Output
137,842
Public Free ($0.00)100%
9
nvidia/nemotron-3-super-120b-a12bnvidia/nemotron-3-super-120b-a12b:free
OpenRouter via OmniRoute
2.9%
7.17M
Input
6,789,259
Output
378,647
Public Free ($0.00)100%
10
antigravity/gemini-3.7-flash-medium
antigravity via OmniRoute
2.5%
6.28M
Input
6,230,219
Output
46,423
Access not reported94.1%
11
antigravity/gemini-3.7-flash-low
antigravity via OmniRoute
2.3%
5.69M
Input
5,538,886
Output
150,225
Public Free ($0.00)84.7%
12
nvidia/google/gemma-4-31b-it
nvidia via OmniRoute
1.9%
4.85M
Input
4,819,219
Output
32,208
Public Free ($0.00)68.7%
13
inclusionai/ling-3.0-flash-fininclusionai/ling-3.0-flash-fin:free
OpenRouter via OmniRoute
1.3%
3.31M
Input
3,204,320
Output
101,756
Public Free ($0.00)100%
14
qwen/qwen3.6-27b
groq via OmniRoute
1.2%
3.07M
Input
2,970,018
Output
101,397
Public Free ($0.00)100%
15
cohere/north-mini-codeopenrouter/cohere/north-mini-code:free
OpenRouter via OmniRoute
1.1%
2.81M
Input
2,805,865
Output
3,841
Public Free ($0.00)—
16
openai/gpt-oss-20b
groq via OmniRoute
1%
2.57M
Input
2,505,095
Output
65,513
Public Free ($0.00)100%
17
cohere/north-mini-codecohere/north-mini-code:free
OpenRouter via OmniRoute
0.9%
2.38M
Input
2,354,110
Output
25,637
Public Free ($0.00)—
18
inclusionai/ling-3.0-flash-vlopenrouter/inclusionai/ling-3.0-flash-vl:free
OpenRouter via OmniRoute
0.9%
2.35M
Input
2,292,519
Output
56,419
Public Free ($0.00)78.4%
19
inclusionai/ling-3.0-flash-finopenrouter/inclusionai/ling-3.0-flash-fin:free
OpenRouter via OmniRoute
0.9%
2.30M
Input
2,260,960
Output
38,099
Public Free ($0.00)62.1%
20
nvidia/nvidia/nemotron-3-super-120b-a12b
nvidia via OmniRoute
0.7%
1.79M
Input
1,638,764
Output
150,347
Public Free ($0.00)58.7%
21
antigravity/gemini-3.1-pro-low
antigravity via OmniRoute
0.7%
1.65M
Input
1,610,259
Output
35,059
Access not reported—
22
gemini-3.7-flash-low
antigravity via OmniRoute
0.7%
1.63M
Input
1,621,480
Output
12,981
Access not reported100%
23
poolside/laguna-xs-2.1poolside/laguna-xs-2.1:free
OpenRouter via OmniRoute
0.6%
1.58M
Input
1,570,780
Output
8,708
Public Free ($0.00)100%
24
nvidia/nemotron-3-super-120b-a12b
nvidia via OmniRoute
0.5%
1.30M
Input
1,256,797
Output
46,084
Public Free ($0.00)100%
25
groq/openai/gpt-oss-120b
groq via OmniRoute
0.5%
1.26M
Input
1,238,962
Output
21,415
Public Free ($0.00)—
26
nex-agi/nex-n2.5-mininex-agi/nex-n2.5-mini:free
OpenRouter via OmniRoute
0.5%
1.25M
Input
1,227,027
Output
23,318
Public Free ($0.00)100%
27
nvidia/nemotron-3-ultra-550b-a55bnvidia/nemotron-3-ultra-550b-a55b:free
OpenRouter via OmniRoute
0.5%
1.13M
Input
1,112,474
Output
19,681
Public Free ($0.00)100%
28
antigravity/claude-sonnet-4-6-high
antigravity via OmniRoute
0.4%
939.9K
Input
927,410
Output
12,509
Public Free ($0.00)—
29
inclusionai/ling-3.0-flash-vlinclusionai/ling-3.0-flash-vl:free
OpenRouter via OmniRoute
0.4%
888.9K
Input
851,655
Output
37,219
Public Free ($0.00)100%
30
claude-sonnet-4-6
antigravity via OmniRoute
0.3%
801.3K
Input
764,482
Output
36,811
Public Free ($0.00)—
31
poolside/laguna-s-2.1poolside/laguna-s-2.1:free
OpenRouter via OmniRoute
0.3%
680.9K
Input
654,688
Output
26,244
Public Free ($0.00)—
32
nvidia/nemotron-3-ultra-550b-a55bopenrouter/nvidia/nemotron-3-ultra-550b-a55b:free
OpenRouter via OmniRoute
0.3%
676.2K
Input
640,341
Output
35,824
Public Free ($0.00)76.7%
33
nvidia/nemotron-3-super-120b-a12bopenrouter/nvidia/nemotron-3-super-120b-a12b:free
OpenRouter via OmniRoute
0.2%
574.4K
Input
463,745
Output
110,623
Public Free ($0.00)—
34
google/gemma-4-31b-it
nvidia via OmniRoute
0.2%
483.0K
Input
471,930
Output
11,117
Public Free ($0.00)—
35
nex-agi/nex-n2.5-pronex-agi/nex-n2.5-pro:free
OpenRouter via OmniRoute
0.2%
466.2K
Input
451,467
Output
14,706
Public Free ($0.00)—
36
gemma4:31b
Local
0.2%
410.0K
Input
397,922
Output
12,094
Local ($0.00)91.4%
37
groq/openai/gpt-oss-20b
groq via OmniRoute
0.2%
395.9K
Input
378,248
Output
17,635
Public Free ($0.00)—
38
nex-agi/nex-n2.5-miniopenrouter/nex-agi/nex-n2.5-mini:free
OpenRouter via OmniRoute
0.2%
380.1K
Input
371,412
Output
8,687
Public Free ($0.00)4.8%
39
gemini-3.1-pro-low
antigravity via OmniRoute
0.1%
349.0K
Input
344,715
Output
4,325
Public Free ($0.00)100%
40
ling-3.0-flash-fin-free
—
0.1%
276.2K
Input
265,856
Output
10,314
Public Free ($0.00)—
41
liquid/lfm-2.5-2.6bliquid/lfm-2.5-2.6b:free
OpenRouter via OmniRoute
0.1%
202.8K
Input
191,701
Output
11,059
Public Free ($0.00)—
42
poolside/laguna-xs-2.1openrouter/poolside/laguna-xs-2.1:free
OpenRouter via OmniRoute
0.1%
190.1K
Input
177,655
Output
12,438
Public Free ($0.00)—
43
openai/gpt-oss-safeguard-20b
groq via OmniRoute
0.1%
152.5K
Input
143,685
Output
8,777
Public Free ($0.00)—
44
inclusionai/ling-3.0-flash-santeinclusionai/ling-3.0-flash-sante:free
OpenRouter via OmniRoute
0.1%
145.2K
Input
135,593
Output
9,619
Public Free ($0.00)—
45
nvidia/nemotron-3-nano-omni-30b-a3b-reasoningnvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free
OpenRouter via OmniRoute
0.1%
144.8K
Input
133,782
Output
11,024
Public Free ($0.00)—
46
claude-sonnet-4-6-high
antigravity via OmniRoute
0%
119.9K
Input
115,158
Output
4,765
Public Free ($0.00)—
47
poolside/laguna-s-2.1openrouter/poolside/laguna-s-2.1:free
OpenRouter via OmniRoute
0%
90.1K
Input
81,608
Output
8,479
Public Free ($0.00)—
48
nvidia/nemotron-3-nano-omni-30b-a3b-reasoningopenrouter/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free
OpenRouter via OmniRoute
0%
82.3K
Input
73,118
Output
9,205
Public Free ($0.00)—
49
nvidia/nemotron-3.5-lightningopenrouter/nvidia/nemotron-3.5-lightning:free
OpenRouter via OmniRoute
0%
82.0K
Input
74,256
Output
7,743
Public Free ($0.00)—
50
liquid/lfm-2.5-2.6bopenrouter/liquid/lfm-2.5-2.6b:free
OpenRouter via OmniRoute
0%
81.3K
Input
76,831
Output
4,434
Public Free ($0.00)—
51
nex-agi/nex-n2.5-proopenrouter/nex-agi/nex-n2.5-pro:free
OpenRouter via OmniRoute
0%
76.9K
Input
75,379
Output
1,481
Public Free ($0.00)—
52
inclusionai/ling-3.0-flash-santeopenrouter/inclusionai/ling-3.0-flash-sante:free
OpenRouter via OmniRoute
0%
57.6K
Input
52,711
Output
4,909
Public Free ($0.00)—
53
nvidia/nemotron-3.5-lightningnvidia/nemotron-3.5-lightning:free
OpenRouter via OmniRoute
0%
29.6K
Input
29,206
Output
386
Public Free ($0.00)—
54
qwen2.5-coder:7b
Local
0%
25.4K
Input
22,502
Output
2,941
Local ($0.00)—
55
auto/codingauto/coding:free
—
0%
19.6K
Input
19,043
Output
528
Public Free ($0.00)—
56
antigravity/gemini-3-flash
antigravity via OmniRoute
0%
13.1K
Input
12,472
Output
625
Public Free ($0.00)—
57
antigravity/gemini-3.7-flash-high
—
0%
8.9K
Input
8,562
Output
380
Public Free ($0.00)—
58
google/gemma-4-26b-a4b-itopenrouter/google/gemma-4-26b-a4b-it:free
OpenRouter via OmniRoute
0%
888
Input
696
Output
192
Public Free ($0.00)0%
59
auto/best-free
—
0%
879
Input
742
Output
137
Public Free ($0.00)—
60
mistralai/mistral-nemotron
—
0%
698
Input
616
Output
82
Public Free ($0.00)—
61
google/gemma-4-31b-itopenrouter/google/gemma-4-31b-it:free
OpenRouter via OmniRoute
0%
666
Input
522
Output
144
Public Free ($0.00)5%
62
google/gemma-4-26b-a4b-itgoogle/gemma-4-26b-a4b-it:free
OpenRouter via OmniRoute
0%
555
Input
435
Output
120
Public Free ($0.00)—
63
google/gemma-4-31b-itgoogle/gemma-4-31b-it:free
—
0%
222
Input
174
Output
48
Public Free ($0.00)—
SWARM COGNITIVE EFFICIENCY

How the work gets done

Explore token usage by agent role, recovery overhead, and request outcomes. Task-stage completion reporting is still to come.

COMMUNITY TELEMETRYReal measurements · published as the swarm grows

More useful measurements, as we grow.

These are the engineering questions we answer with real software tasks: recovery overhead, tokens by role, provider reliability, and multi-model synergy.

New detailed stats · partial history. These panels cover only requests with the required signed details, so their token counts will not match the lifetime totals elsewhere on the site. Coverage grows as new reports arrive and older details are verified.

Regular work versus requests marked as repair or recovery in this reporting sample.

Regular work: 91.9%
Recovery: 8.1%
Regular work81,317,695 tokensRequests not marked as recovery
Recovery & Repair Compute7,170,473 tokensReported retries and check fixes in this sample
Repair Overhead8.1%Share of this sample’s tokens spent on repair
Run cheapoS on your own code.Autonomous software loops with zero-dollar developer quotas, local Ollama, or frontier keys. It's open source and ready to run.
Get cheapoS on GitHub ↗How the Club Works →
INFRASTRUCTURE PROVENANCE

Where Community Compute Originates

4 compute tiers working together: public free gateways, included developer quotas, local hardware, and pay-as-you-go frontier keys.

99.2%

Public Free Gateways

Tokens reported under public free access. Availability and quotas depend on the provider.

248,713,832 tokens
0.6%

Included Account Quotas

Tokens reported under included access. The account or subscription may have a separate cost.

1,601,645 tokens
0.2%

Local Hardware

Tokens reported as local inference. Hardware and electricity costs are separate.

435,459 tokens
0.0%

Paid Pay-As-You-Go

Tokens reported under paid API access.

0 tokens