Meituan · Chat and reasoning · 69.1B parameters · mit
LongCat Flash Lite
Needs 83 GB at FP8 / INT8. The smallest real configuration today is 4× RTX 3090 at Clore.ai, $0.25 per hour, about $183 a month at 730 hours.
Parameters69.1Bdense
Context320k tokensmaximum published
Checkpoint138 GBBF16, as published
Licensemitcheck before commercial use
Released—documented date
Smallest configuration · FP8 / INT8
4× RTX 309096 GB VRAM · 86% used
- Per hour
- $0.25
- Per month
- $183
- Cheapest at
Clore.ai
Memory · 3 precisions · 6 context lengthsHow much VRAM it needs
Weights plus runtime overhead, then the KV cache that grows with the context you serve.
| Precision | Weights | Weights + headroom | 4k context | 32k context | 128k context | 256k context | 320k context |
|---|---|---|---|---|---|---|---|
| BF16published checkpoint | 138 GB | 166 GB | 166 GB | 167 GB | 168 GB | 171 GB | 172 GB |
| FP8 / INT8estimated | 69 GB | 83 GB | 83 GB | 84 GB | 86 GB | 88 GB | 89 GB |
| INT4 / MXFP4estimated | 41 GB | 50 GB | 50 GB | 51 GB | 52 GB | 54 GB | 56 GB |
| KV cacheone sequence, FP16 cache, included above | — | — | 1 GB | 1 GB | 3 GB | 5 GB | 6 GB |
Cross-checked on Hugging Face Oct 28, 2026: 69.1B parameters in the safetensors headers, 138.2 GB of weight files, attention layout from config.json (14 layers, latent rank 512). Method: weights as published at the native precision, verified on Hugging Face 2026-10-28; other precisions estimated at 1 (8-bit) and 0.6 (4-bit) bytes per parameter; × 1.2 runtime headroom; KV cache = 1152 bytes per token per layer × 14 full-attention layers (FP16 cache, one sequence).config.json ↗
Smallest configuration per GPU · 110 GPUs · cheapest eligible on-demand offer per GPU count, every provider on fileWhat it costs to run, GPU by GPU
The fewest GPUs of one type whose total memory fits the model at short context, priced from the cheapest eligible on-demand offer for exactly that count (45-day window, newest reading per provider), with the next providers beside it. Monthly = 730 hours.
| GPU | Smallest config | VRAM used | $ / hour | $ / month | Cheapest at |
|---|---|---|---|---|---|
| RTX 3090Best fit | 4× 24 GB96 GB total | 86% | $0.25 | $183 | Clore.aiTrust 46 · read Nov 2 · receipt · also Vast.ai $0.56, Edgevana $0.92 |
| Tesla V100 32GB | 4× 32 GB128 GB total | 65% | $0.27 | $196 |  Vast.aiTrust 78 · read Nov 2 · receipt · also Cyfuture AI $2.22, gcube $6.44 |
| RTX A6000 | 2× 48 GB96 GB total | 86% | $0.38 | $274 | Clore.aiTrust 46 · read Nov 2 · receipt · also Lium $0.84, Spheron $0.89 |
| Tesla P100 | 8× 16 GB128 GB total | 65% | $0.38 | $274 | Vast.aiTrust 78 · read Nov 2 · receipt |
| RTX 3060 | 8× 12 GB96 GB total | 86% | $0.43 | $312 | Vast.aiTrust 78 · read Nov 2 · receipt · also Edgevana $0.64, AtmosCompute $1.01 |
| Quadro RTX 8000 | 2× 48 GB96 GB total | 86% | $0.48 | $354 | Vast.aiTrust 78 · read Nov 2 · receipt · also Edgevana $0.56 |
| L40 | 2× 48 GB96 GB total | 86% | $0.54 | $391 | Vast.aiTrust 78 · read Nov 2 · receipt · also io.net $1.32, GPU.ai $1.56 |
| RTX 2080 Ti | 8× 11 GB88 GB total | 94% tight | $0.59 | $432 | Vast.aiTrust 78 · read Nov 2 · receipt · also Edgevana $0.68 |
| RTX PRO 6000 | 1× 96 GB96 GB total | 86% | $0.66 | $482 |  packet.ainot scored · read Nov 2 · receipt · also Runpod $0.69, HyperAI $0.80 |
| Titan RTX | 4× 24 GB96 GB total | 86% | $0.70 | $508 | Vast.aiTrust 78 · read Nov 2 · receipt |
| Tesla V100 16GB | 8× 16 GB128 GB total | 65% | $0.75 | $547 | Vast.aiTrust 78 · read Nov 2 · receipt · also CloudRift $2.32, Lambda $6.32 |
| L40S | 2× 48 GB96 GB total | 86% | $0.76 | $555 |  LiumTrust 40 · read Nov 2 · receipt · also Vast.ai $1.07, CloudRift $1.26 |
| RTX A5000 | 4× 24 GB96 GB total | 86% | $0.80 | $586 | Vast.aiTrust 78 · read Nov 2 · receipt · also Edgevana $1.05 |
| RTX 5060 Ti | 8× 16 GB128 GB total | 65% | $0.82 | $596 | Clore.aiTrust 46 · read Nov 2 · receipt · also Vast.ai $0.86, Edgevana $0.98 |
| RTX PRO 4000 | 4× 24 GB96 GB total | 86% | $0.83 | $608 | Clore.aiTrust 46 · read Nov 2 · receipt · also Vast.ai $1.55, Edgevana $1.78 |
| RTX 4070 Super | 8× 12 GB96 GB total | 86% | $0.85 | $617 | Vast.aiTrust 78 · read Nov 2 · receipt · also AtmosCompute $1.36 |
| A40 | 2× 48 GB96 GB total | 86% | $0.94 | $683 | Vast.aiTrust 78 · read Nov 2 · receipt · also Exoscale $2.09 |
| A800 80GB PCIe | 2× 80 GB160 GB total | 52% | $0.96 | $702 | Vast.aiTrust 78 · read Nov 2 · receipt |
| RTX PRO 5000 | 2× 48 GB96 GB total | 86% | $0.96 | $702 | Vast.aiTrust 78 · read Nov 2 · receipt · also Lium $1.70, AtmosCompute $2.04 |
| A10 | 4× 24 GB96 GB total | 86% | $0.96 | $703 | Vast.aiTrust 78 · read Nov 2 · receipt · also Volcano Engine $6.39 |
| RTX A4000 | 8× 16 GB128 GB total | 65% | $0.96 | $704 | Vast.aiTrust 78 · read Nov 2 · receipt |
| RTX 6000 Ada | 2× 48 GB96 GB total | 86% | $0.99 | $722 | Vast.aiTrust 78 · read Nov 2 · receipt · also GPU.ai $1.30, io.net $1.37 |
| Gaudi 2 | 1× 96 GB96 GB total | 86% | $1.06 | $777 |  Cyfuture AInot scored · read Nov 2 · receipt |
| RTX 4090 | 4× 24 GB96 GB total | 86% | $1.08 | $791 | Clore.aiTrust 46 · read Nov 2 · receipt · also Vast.ai $1.50, Lium $2.00 |
| The 24 cheapest of 61 GPU types that fit. Every GPU, any precision and context, in the finder | |||||
Throughput estimates · Oct 4, 2026Tokens per second, and what a million costs
Chat and reasoning. Speed estimated from memory bandwidth and model size, never measured. The cost per 1M output tokens is an estimate at the prices read on Oct 4, 2026; it is not recomputed here.
| GPU | Cards | Precision | tok/s per user | per 1M output tokens · estimate |
|---|---|---|---|---|
| MI300X | 1 | FP8 | 40.6 | $0.428 |
| RTX 4090 | 4 | FP8 | 25.1 | $0.500 |
| B200 | 1 | FP8 | 73.8 | $0.539 |
| H100 SXM | 2 | FP8 | 56.9 | $0.605 |
| H200 | 1 | FP8 | 45.8 | $0.637 |
| RTX PRO 6000 | 1 | FP8 | 17.8 | $0.645 |
| L40S | 2 | FP8 | 13.6 | $0.652 |
| H100 PCIe | 2 | FP8 | 30.2 | $0.746 |
| RTX 5090 | 4 | FP8 | 42.8 | $0.764 |
| A100 80GB | 2 | BF16 | 18.9 | $1.54 |
cards, precision and speed estimated from memory bandwidth and model size; cost recomputed by the site from verified prices · cheapest per output at the prices of Oct 4, 2026: MI300XWhy estimates, and their limits →
LongCat Flash Lite · related