DeepSeek R1 Distill Llama 70B API pricing

9 providers serve DeepSeek R1 Distill Llama 70B. DeepSeek R1 Distill Llama 70B is an open-weight model (MIT) with 70B total parameters, up to 131K context. Prices below are per 1M tokens, cheapest input first.

Refreshed 2026-11-22 · list prices from provider APIs
Input floor $0.200/1M at DeepInfra Output floor $0.375/1M at Nscale Find your workload winner →
Provider Input $/1M Output $/1M Cache read $/1M Context
DeepInfra $0.200 $0.600 - 131K Use
Nebius $0.250 $0.750 - 128K Use
GMI Cloud $0.250 $0.750 - 131K Use
Nscale $0.375 $0.375output floor - - Use
OVHcloud $0.670 $0.670 - 131K Use
Vercel AI Gateway $0.750 $0.990 - 131K Use
Novita AI $0.800 $0.800 - 8K Use
Fireworks AI $0.900 $0.900 - 131K Use
DigitalOcean (Paperspace) $0.990 $0.990 - 33K Use

The cheapest way to run DeepSeek R1 Distill Llama 70B

On an illustrative 70M-input / 30M-output monthly workload, today's lowest listed cost is roughly $32.00/month, at DeepInfra. Input-heavy, output-heavy and cache-heavy workloads can produce different winners. Use the live mix calculator below rather than combining floors from two different providers. On output-token cost alone, the API floor remains below our self-hosting estimate of ~$0.480/1M. Details below. Run your own numbers in the breakeven calculator, or read the breakeven math.

Your actual monthly cost

Uses today's list prices

Cache read uses the listed cache rate where available; otherwise it falls back to normal input price. Batch, write-cache, volume and negotiated discounts are excluded.

Or run it yourself: DeepSeek R1 Distill Llama 70B self-hosting economics

DeepSeek R1 Distill Llama 70B is open-weight (MIT), so the API price above competes with the GPU-hour market. 70B total parameters needs roughly 97 GB VRAM at FP8 or 49 GB at INT4, KV-cache headroom included.

GPU that fits (FP8, single card)VRAMFrom $/hrCheapest at
MI300X 192 GB $0.500community RunPod Rent
MI350 OAM 294 GB $0.500community RunPod Rent
H200 NVL 141 GB $0.500community RunPod Rent
H200 141 GB $3.52spot Verda
B200 180 GB $4.87spot Verda
B300 288 GB $6.53spot Verda
Breakeven estimate: a well-batched vLLM deployment on one MI300X ($0.500/hr) at ~579 aggregate tok/s and 50% utilization works out to roughly $0.480 per 1M output tokens, versus $0.375 via the cheapest output-token API (Nscale). The API is cheaper unless your utilization or throughput beats these assumptions. Planning estimate: throughput varies with hardware, quantization, batch size and context. Tune it in the calculator.

Related models

ModelCheapest in $/1MCheapest out $/1MProviders
DeepSeek V4 Flash $0.044 $0.149 31
DeepSeek V4.1 Flash $0.010 $0.199 26
DeepSeek V4 Pro $0.528 $1.20 24
DeepSeek R1 $0.200 $0.400 20
DeepSeek V3 $0.190 $0.200 18
DeepSeek-V3.1 $0.250 $0.950 13

FAQ

What is the cheapest API for DeepSeek R1 Distill Llama 70B?

As of 2026-11-22, the lowest input price for DeepSeek R1 Distill Llama 70B is DeepInfra at $0.200 per 1M input tokens. The lowest output price is Nscale at $0.375 per 1M output tokens. The cheapest provider for a real workload depends on its input/output mix. The next-lowest input price is Nebius at $0.250, a 25% difference.

How much VRAM do you need to self-host DeepSeek R1 Distill Llama 70B?

DeepSeek R1 Distill Llama 70B has 70B parameters, so plan for roughly 97 GB of VRAM at FP8 or 49 GB at INT4/AWQ, KV-cache headroom included. A single MI300X (192 GB) fits it.

← All models · GPU rental prices · Breakeven calculator · Get this data as JSON