Log in Become a host
B300 288GB hosts wanted — $4.77–$5.80 / hr / card

Got a B300?
Put it to work.

List your B300 on Clore.ai. Blackwell Ultra supply is the tightest constraint of 2026 — every properly-configured 8-GPU NVLink-Switch pod gets noticed by labs blocked on hyperscaler quota gates. Net around $3,091/month per card before MFP staking, paid per-minute in BTC, USDT, USDC or CLORE. Highest sustained price tier on the platform. Stake CLORE for up to +200% daily emission.

BTC USDT USDC CLORE
Per-minute billing Withdraw any time You set the price
DISTRIBUTION STATEMENT · Q2 2026 A/C · CLR-OP-B300-04829
PERIOD 01 APR — 30 JUN 2026
HOLDING B300 SXM5 ×8 · NVLink-Switch
DEPLOYED 214 days · 99.94% uptime
MTD
$24,728
▲ 4.1%
QTD
$71,200
▲ 6.8%
YTD
$278,190
▲ 12.4%
List rate · 8× pod$38.16 / hr
Spot fee (base)−2.5%
Proof-of-Holding rebate−1.8%
Renter-pays-half→ 1.25%
Net distribution / month $24,728USD
+ CLORE bonus distribution +$12,365USD value
Per hour
$34.34
Per day
$824.27
Total max
$37,093/mo
$3091
Avg. monthly rental net per B300
1.8%
Lowest spot fee with PoH (50% reduction)
+50%
Up to — CLORE bonus on top of every rental.
1min
Per-minute billing — earnings credit every 60 s.
why host on clore

Idle GPUs are
idle revenue.

If you've got a B300 — or a whole rack of them — you've already paid for the silicon, the power contract, the rack space. Clore turns that capex into per-minute revenue in your choice of crypto, no sales calls, no minimum commitment.

The 288 GB memory premium

Frontier labs willing to pay the B300 premium are the ones training 1T+ parameter models or serving 671B MoE in production — workloads that B200 can't sustain. NVLink-Switch fabric pods reliably clear the top of the marketplace price range.

Spot fee2.5% → 1.8%
On-demand fee10% → 5%

Per-minute payouts

Earnings credit to your wallet balance every minute the rental runs. Withdraw to BTC, USDT, USDC or CLORE as often as you want — no daily caps.

Billing cadencePer minute
Withdraw capNone

One card or a rack

One B300 in a closet, or up to 192 servers per account onboarded via API. Same console, same fees, same flow.

Min servers1
Default cap192
★ OPTIONAL · BIGGER PAYOUTS

Hold CLORE,
earn ~50% more.

Stake some CLORE behind your machine and the network pays you a daily bonus on top of every rental — about half of what the renter pays you, again in CLORE. Skip it and you still get paid the normal way; this just stacks more on.

RENTAL

What renters pay you

100%of your list price

Per-minute, in BTC / USDT / USDC / CLORE. Withdraw any time.

+ BONUS (OPTIONAL)

Network reward in CLORE

+50%of rental, paid in CLORE

Activated by staking CLORE behind your machine. Paid daily, on top of normal earnings.

Optional — no penalty if you skip · Paid daily in CLORE · Read how the bonus works →
fees

Two fees. That's it.

Clore takes a small cut of every rental. Half is paid by you, half by the renter — so the number below is the full marketplace fee, not what comes out of your pocket.

Spot orders
2.5%

Interruptible — best for mining & batch jobs

You pay 1.25%, the renter pays the other 1.25%. Hold CLORE to cut your share even more.

On-demand orders
10%

Reserved — best for AI & production

You pay 5%, the renter pays the other 5%. Hold CLORE to cut your share even more.

// Want the full breakdown of fee reductions and edge cases? Read the fee docs →

what you'll earn

Three ways to host.
Pick yours.

All numbers below assume a B300 listed at $5.80/hr. Real numbers depend on demand, your price, and your power costs.

Easy mode
$3091

Just plug it in

List your card, accept rentals. No CLORE required, no setup beyond the host software. Get paid per minute.

+ Hold CLORE
$3101

Lower fees

Hold CLORE in your wallet — no lock, no contract. Your half of the marketplace fee drops by up to 50%.

+ Stake CLORE for the bonus
up to $4,636

Maximum payout

Stake CLORE behind your server to unlock a daily network bonus on top of your rental — adds about half your rental income again, paid in CLORE.

// Numbers are a rough monthly estimate, not a guarantee — bonus depends on competition. Full host guide →

how to host

From rig to revenue
in under 30 minutes.

A Linux box, the Clore hosting software, a stable internet connection. One B300 or a tier-3 facility — same flow.

01 / INSTALL

Install Clore Hosting Software

Boot from the Clore Linux image (USB or PXE). Pair the host with your account using your initialization token.

02 / CONFIGURE

Set server visibility

Configure SSH, Docker, and per-card settings. Flip the server to public when it passes the auto-attestation.

03 / PRICE

Set spot & on-demand

List both, or just one. Adjust live — the floor for B300 288GB spot is $4.77 / hr right now.

04 / BOOST

Lock MFP (optional)

Lock 52,000 CLORE for Tier 1 (or up to 260,000 for Tier 2). 24 h warm-up, then up to +200% rental price as daily rewards.

faq

Questions hosts ask.

How is B300 different from B200?

B300 is Blackwell Ultra — same FP4/FP8 tensor cores as B200 but with 288 GB HBM3e (50% more memory) and higher sustained throughput. The extra VRAM unlocks 1T+ parameter pretraining and 671B MoE serving with full KV cache headroom, where B200 had to swap experts. CLORE.AI lists B300 as supply ramps in 2026; available on bare-metal 8-GPU NVLink-Switch pods.

What infrastructure do I need to host H100, H200, or B200?

Datacenter-class power (208/415V three-phase recommended for 8-GPU SXM pods at 5.6 kW for H100/H200 and 8 kW for B200), liquid cooling for B200 SXM, redundant 100 Gbps networking, and a host that can saturate it. PCIe variants run on dual-socket x86 servers with 1500-2000W PSUs. Most CLORE.AI datacenter hosts deploy these in colo cages, not home racks.

Are there enterprise SLAs and fleet contracts available?

CLORE.AI is a marketplace - SLAs are negotiated host-to-renter through on-demand reserved listings, not centrally enforced. Operators running 16+ datacenter GPUs commonly offer 99.9% uptime contracts directly to enterprise renters. Talk to the platform team for fleet onboarding tooling and bulk-listing API access if you're deploying 32+ cards.

How does the +200% MFP bonus scale on a $1,000+/month card?

MFP daily emission reward is up to +200% of the server's rental price, paid in CLORE on top of fiat-equivalent rental income. On an H100 earning $1,170/month, the maximum MFP bonus tracks rental revenue directly - so the absolute bonus on datacenter cards dwarfs consumer-tier numbers. Stake CLORE behind your fleet at the same ratio across all servers; the bonus also zeroes the non-CLORE hoster fee.

Do training renters expect 99.9% uptime?

Yes. Multi-week pretraining runs on 8-GPU pods are economically catastrophic to interrupt - a checkpoint loss on a 7-day H100 run costs the renter low-five-figures in lost compute. Datacenter hosts with 99.9%+ verified 30-day uptime command premium pricing and clear fleet-scale demand. Reliability score on your listing is the primary signal training renters filter by.

Is power and cooling the binding constraint?

For B200 SXM, yes - 1,000 W TDP per GPU, 8 kW per node, liquid-cooled rear-door heat exchangers or direct-to-chip loops are effectively required. H100 and H200 air-cool in well-designed colo (700 W TDP each), but rack density lands 2-4 GPUs per U at most. A100 (400 W) and earlier are easier - any datacenter with 30 kW/rack handles them comfortably.

workload spotlight

Real numbers on the B300.

288 GB HBM3e + Blackwell Ultra FP4 — the trillion-parameter inference card. Pretrains 1T+ models and serves 671B MoE without expert swap.

Pretrain 1T+ MoE (8-GPU pod)
TransformerEngine FP4 + 5th-gen NVLink + FSDP
~16,000 tokens/s/GPU on 8× node

288 GB per card carries 1T+ parameter active weights without B200's expert-swap overhead. Frontier-lab unlock for trillion-scale pretraining.

Read the guide →
DeepSeek-V3 671B MoE serving
vLLM + expert parallelism + FP8/FP4
~5,200 tok/s aggregated on 8× node

288 GB per card serves 671B MoE with full KV cache headroom — no off-card expert spill, ~37% higher sustained throughput than B200.

Read the guide →
Llama-3 70B FP4 serving
vLLM + Blackwell Ultra FP4 + chunked prefill
~7,500 tok/s aggregated, 256 concurrent

Doubles B200 70B concurrency at the same per-request latency budget. The serving-throughput ceiling in 2026.

Read the guide →
datacenter comparison

Datacenter-tier comparison.

Side-by-side specs across the datacenter tier. Click any row to see that GPU.

GPU
HBM
Mem BW (GB/s)
FP8 TFLOPS
BF16 TFLOPS
NVLink BW
Transformer Engine
Spot $/hr
Tesla V100
32 GB HBM2
900
300 GB/s
$0.28
A100 40GB
40 GB HBM2e
1,555
312
600 GB/s
$0.78
A100 80GB
80 GB HBM2e
1,935
312
600 GB/s
$0.92
H100 tier focus
80 GB HBM3
3,350
1,979
989
900 GB/s
yes
$1.89
H200
141 GB HBM3e
4,800
1,979
989
900 GB/s
yes
$2.40
B200
192 GB HBM3e
8,000
~4,500
~2,250
1,800 GB/s
yes
$3.40
host docs

Everything you need to host an B300.

Read the host onboarding guide, MFP staking mechanics, and marketplace fee schedule.

Host docs
Becoming a host
Host docs
MFP — Maximum Fair Price
Host docs
Marketplace fees
Host docs
All docs
other gpus

Compare with similar cards.

B200
192 GB · earn ~$2589/mo
Host →
H200
141 GB · earn ~$1185/mo
Host →
H100
80 GB · earn ~$1170/mo
Host →

Your B300 starts
earning tonight.

List your card, pick spot or on-demand, and start collecting per-minute earnings in BTC, USDT, USDC, or CLORE. Lock MFP whenever you're ready.