P4 of 22 ▲17 · 166 pts · +3304 this week
online nowclaimedThe Dalles, Oregon1 machineearliest result held 2026-11-04
Points
166
2557 behind P1
Signed runs
34
17 in 30 days
Builds measured
34
Best results held
14
3 pts each
Firsts
9
10 pts each
Rooms
0
Carbon per M tokens
1.4 kg
median · modelled · United States grid average
Gram ·662a4e is P4 of 22 on the 50GRAMx provider standings with 166 points, from 34 signed runs on 1 machine in The Dalles, Oregon. Their fastest signed result is Llama 3.2 1B at 11 tokens per second. They hold the network's best result on 14 builds and were the first machine anywhere to measure 9. Every number is signed by the machine that measured it.
Fastest models on their machines
Output tokens per second, signed · higher is better
Llama 3.2 1B11 tok/s
Qwen 3 VL-2B9 tok/s4-bit by Qwen
gpt-oss 20B5 tok/s4-bit by unsloth
OpenAI-gpt-oss-20B-Claude-4.5-Opus-Heretic-Uncensored4 tok/s1-bit by mradermacherQwen 3.5 35B-A3B4 tok/s4-bit by unsloth
GLM-4.7 Flash3 tok/s4-bit by unsloth
North 1.0 Mini-Code-1-02 tok/s3-bit by mradermacher
Devstral-Small-2505-Deepseek-V3.2-Speciale-Distill0 tok/s3-bit by TeichAIA million tokens here
Metered on their machine · OpenRouter price beside it where one is sold
Llama 3.2 1B$2.82OpenRouter $0.20Qwen 3 VL-2B$3.38no API sells itgpt-oss 20B$8.63OpenRouter $0.13OpenAI-gpt-oss-20B-Claude-4.5-Opus-Heretic-Uncensored$9.39no API sells itQwen 3.5 35B-A3B$12OpenRouter $1.00GLM-4.7 Flash$14OpenRouter $0.40North 1.0 Mini-Code-1-0$18no API sells itDevstral-Small-2505-Deepseek-V3.2-Speciale-Distill$105no API sells itWhat they measured, by family
Signed runs per model family
Every part as the gram benchmarked and signed it.
CPU only
linux · x86_64
CPU
Intel Xeon @ 2.2 GHz
2 cores / 4 threads · 55 MB L3 · 467 hashes/s all cores
Memory
31.3 GB
2 GB/s · 86.4 ns latency
Disk
overlayfs
2585 MB/s read · 172 MB/s write · 262855 IOPS
Network
8.3 MiB/s download
Where
The Dalles, Oregon
Against every machine on the network
Linux · no GPU · 4 CPUs · 31 GB RAM. Every part is benchmarked and signed by the machine itself.
34 signed runs across 34 builds; 17 in the last 30 days.
Llama 3.2 1B at 11 tokens per second, signed.
1 point per signed run, 3 per build where the provider holds the network’s fastest result on that kind of machine, 10 per model they were first to measure.
Yes. Each run publishes the exact bytes the machine signed and its signature; “verify this yourself” checks it in your browser against the machine’s own key.
The seasons table, narrowed to this provider’s signed runs: speed and cost on their machines, next to what OpenRouter charges, the checks each model passed, and its licence.
| parameters | tokens/sec | proved | per M tokens | per M tokens | Verdict | ||||
|---|---|---|---|---|---|---|---|---|---|
Text & chat · by openai8 builds · 5 variants | 20B | 10.8 GB | 4.64 CPUs, no GPU· Kaggle | 128k | 8 + 18 lang | $8.63metered | $0.1311 providers | Anyone | API 66× cheaper |
Text & chat · by zai-org3 builds · 1 variant | ≈28B | 15.7 GB | 3.24 CPUs, no GPU·Kaggle | 128k | 8 + 16 lang | $14metered | $0.402 providers | Anyone | API 35× cheaper |
Text & chat · by Qwen7 builds · 3 variants | 35B3B active | 19 GB | ran, not timed | 4k | 8 + 9 lang | — | $0.9011 providers | Anyone | measured |
Text & chat · by CohereLabs5 builds | ≈31B | 17.2 GB | 2.14 CPUs, no GPU·Kaggle | 4k | 7 + 9 lang | $18metered | not sold | Anyone | only on this network |
Text & chat · by Qwen4 builds | 35B3B active | 20.5 GB | 4.04 CPUs, no GPU·Kaggle | 4k | 6 + 14 lang | $12metered | $1.007 providers | Anyone | API 12× cheaper |
Text & chat · by meta-llama10 builds | 1.2B | 0.8 GB | 114 CPUs, no GPU·Kaggle | 32k | 5 + 8 lang | $2.82metered | $0.201 provider | Anyone | API 14× cheaper |
Text & chat · by Qwen1 build | 2B | 1 GB | 9.24 CPUs, no GPU·Kaggle | 4k | 4 | $3.38metered | not sold | Anyone | only on this network |
Text & chat · by TeichAI1 build | ≈21B | 9.2 GB | 0.44 CPUs, no GPU·Kaggle | 4k | 4 + 6 lang | $105metered | not sold | Anyone | only on this network |
Text to speech · by hexgrad1 build | — | 0.3 GB | ran, not timed | — | 1 + 1 lang | — | not sold | Not checked | only on this network |
Speech to text · by UsefulSensors1 build | — | 0.2 GB | ran, not timed | — | 1 + 1 lang | — | not sold | Anyone | only on this network |
Speech to text · by UsefulSensors1 build | — | 0.1 GB | ran, not timed | — | 1 + 1 lang | — | not sold | Anyone | only on this network |
 Music & sound · by facebook1 build | 300M | 2.2 GB | ran, not timed | — | 1 | — | not sold | Own machines only | only on this network |
Vision (counting, locating) · by facebook1 build | 90M | 0.7 GB | ran, not timed | — | 1 | — | not sold | Anyone | only on this network |
 Image generation · by stabilityai2 builds | 900M | 2.4 GB | ran, not timed | — | 1 | — | not sold | Not checked | only on this network |
 Other · by ununununununununununununununununununun1 build | — | — | ran, not timed | — | 1 | — | not sold | You & your company | only on this network |
 Speaker recognition · by speechbrain1 build | 20M | 0.1 GB | ran, not timed | — | 1 | — | not sold | Anyone | only on this network |
 Image generation · by segmind1 build | 320M | 1 GB | ran, not timed | — | 1 | — | not sold | Anyone | only on this network |
Speech to text · by openai1 build | 810M | 1.5 GB | ran, not timed | — | 1 + 10 lang | — | not sold | Anyone | only on this network |
 Vision (counting, locating) · by depth-anything1 build | 100M | 0.4 GB | ran, not timed | — | — | — | not sold | Own machines only | only on this network |
Vision (counting, locating) · by depth-anything1 build | 30M | 0.1 GB | ran, not timed | — | — | — | not sold | Anyone | only on this network |
Translation · by facebook1 build | 600M | 2.3 GB | ran, not timed | — | 0 + 12 lang | — | not sold | Own machines only | only on this network |
