Skip to content

Cost reference

What it costs to fine-tune Qwen3 32B

The cheapest setup we compared is LoRA on 2 × L40S: about $0.27 per million training tokens, so $2.74 for 10 million and $27.40 for 100 million. Full fine-tuning costs about $4.85 per 10 million tokens on 4 × B200.

By Will Schulz · Checked 15 November 2026 · Prices as of November 2026

GPU-hours and cost by GPU and method

Per epoch over the token count shown; multiply by your number of epochs. GPU-hours are the same however many GPUs the job is split over.

GPU · methodGPUs for memoryTokens/s per GPUGPU-hours per 1M1.0M tokens10M tokens100M tokens
H100 SXM · LoRA$4.39/hr23,0190.09$0.40$4.04$40.40
H100 SXM · QLoRA$4.39/hr12,3550.12$0.52$5.18$51.79
H100 SXM · Full fine-tuning$4.39/hr82,0120.14$0.61$6.06$60.59
B200 · LoRA$7.99/hr16,8680.04$0.32$3.23$32.32
B200 · QLoRA$7.99/hr15,3570.05$0.41$4.14$41.43
B200 · Full fine-tuning$7.99/hr44,5780.06$0.48$4.85$48.48
L40S · LoRA$1.09/hr21,1050.25$0.27$2.74$27.40
L40S · QLoRA$1.09/hr18620.32$0.35$3.51$35.13
L40S · Full fine-tuning$1.09/hr13 (over 8)7370.38———
RTX 5090 · LoRA$1.39/hrDoesn't fit6390.43———
RTX 5090 · QLoRA$1.39/hr14990.56$0.77$7.74$77.41
RTX 5090 · Full fine-tuning$1.39/hrDoesn't fit4260.65———

Compared with fine-tuning APIs

OptionPer 1M tokens10M tokens100M tokens
Rented GPU: LoRA on L40S (our estimate)$0.27$2.74$27.40
Together AIQwen3 32B not listed——
OpenAI gpt-4.1-mini fine-tuning (closed model; closed to new users)$5.00$50.00$500

API prices are per training token processed (dataset tokens × epochs), seen on 15 November 2026. Together AI charges a minimum per job. OpenAI's pricing page says fine-tuning is closed to new users. Rented GPU estimates leave out setup, failed runs and idle time, which an API absorbs.

Your run

Pre-filled for Qwen3 32B: 10M tokens, 3 epochs, LoRA on an H100.

Estimated cost

$12.12

H100 SXM-hours

2.8

Time on 2 GPUs

1.4 hours

Tokens per second per GPU

3,019

30M tokens processed. Needs 2 × H100 SXM for memory at 2,048-token sequences. GPU-hours assume perfect scaling across GPUs.

Assumptions

  • Compute per token: 6 FLOPs per active parameter for full fine-tuning (forward plus backward) and about 4 for LoRA and QLoRA, which skip the frozen weights' gradients. Qwen3 32B uses 32.762B parameters per token.
  • Utilisation: 40% of each GPU's dense BF16 peak. torchtune measured 3,083 tokens a second for Llama 3.1 8B LoRA on one RTX 4090, about 60% on the same accounting, so well-tuned code can beat these figures.
  • QLoRA runs at 78% of LoRA's speed, the ratio torchtune measured (2,413 vs 3,083 tokens a second), because the 4-bit weights are dequantised on the fly.
  • Memory decides the GPU count, from our Qwen3 32B VRAM estimates at 2,048-token sequences. Consumer cards are counted as single GPUs only.
  • Prices: on-demand Runpod Secure Cloud rates (page dated 12 November 2026). Community Cloud and spot capacity cost less.

Frequently asked questions

How much does it cost to fine-tune Qwen3 32B?
About $0.27 per million training tokens at the cheapest setting we compared (LoRA on 2 × L40S, $1.09/hour per GPU), so about $2.74 for 10M tokens and $27.40 for 100M. Budget for several runs: the first rarely wins.
How long does it take to fine-tune Qwen3 32B on an H100?
LoRA processes about 3,019 tokens a second on one H100 at 40% utilisation, so 10M tokens take about 55 minutes of GPU time across 2 H100s. Full fine-tuning needs 8 H100s for memory and about 1.5× the compute.
Is it cheaper to fine-tune Qwen3 32B through an API?
Together AI's November 2026 price list doesn't include Qwen3 32B, so the choice is rented GPUs or a hosted model. Rented GPU time costs about $2.74 per 10M tokens by our estimate, plus setup and idle time.
Should I use LoRA or full fine-tuning for Qwen3 32B?
Start with LoRA or QLoRA. Full fine-tuning Qwen3 32B needs about 529 GB, so 8 × 80 GB GPUs with sharded optimiser state, and about 1.5× the compute per token. LoRA usually gets close for format, style and task adaptation; move to full fine-tuning only when evaluation shows LoRA has plateaued.

Sources

  1. [1]Hugging Face. Qwen/Qwen3-32B config.json
  2. [2]Alibaba Qwen. Qwen/Qwen3-32B model card
  3. [3]Alibaba Qwen. Qwen3 Technical Report (arXiv:2505.09388)
  4. [4]Hu et al., 2021. LoRA: Low-Rank Adaptation of Large Language Models (arXiv:2106.09685)
  5. [5]Dettmers et al., 2023. QLoRA: Efficient Finetuning of Quantized LLMs (arXiv:2305.14314)
  6. [6]Kaplan et al., 2020. Scaling Laws for Neural Language Models (arXiv:2001.08361)
  7. [7]PyTorch. torchtune: memory and throughput benchmarks (Llama 3.1 8B on RTX 4090)
  8. [8]Together AI. Fine-tuning pricing (seen 15 November 2026)
  9. [9]OpenAI. API pricing: fine-tuning (seen 15 November 2026)
  10. [10]Runpod. GPU pricing (updated 12 November 2026)

Try the cheaper levers first.

Prompt tuning and model search in Kojen cost API calls, not GPU-hours, and often close the gap before any fine-tuning.

Start training free