LLM Pricing
Back to articles

Data-led article

Cheapest LLM API models for coding workloads

Coding workloads have longer prompts, more context, and larger outputs than a support chatbot. This screen prices a single-turn coding task: 2,000 input tokens, 1,500 output tokens, 50 tasks per user, and 100 monthly users.

Updated June 20, 2026 2,000 input / 1,500 output tokens 5,000 monthly tasks

Scenario

The coding workload being priced

The numbers are generated from the site's current pricing data. Provider bills can differ because of cache behavior, discounts, regions, tiers, or provider-specific billing rules.

Input / task
2,000 tokens

Prompt plus repo context or file snippets.

Output / task
1,500 tokens

Code generation, explanation, or refactoring output.

Tasks / user
50

Repeated coding tasks per month.

Monthly users
100

5,000 total tasks.

Cost screen

Lowest-cost priced routes in this coding workload

Rows are sorted by estimated monthly API cost. Open each model page before treating any route as production-ready.

Model Context Input / 1M Output / 1M Monthly cost
Qwen2.5-Coder-3B-Instruct
nscale
 N/A $0.0100 / 1M tokens $0.0300 / 1M tokens $0.33
Qwen2.5-Coder-7B-Instruct
nscale
 N/A $0.0100 / 1M tokens $0.0300 / 1M tokens $0.33
Qwen2.5-Coder-7B
nebius
 32.8K $0.0100 / 1M tokens $0.0300 / 1M tokens $0.33
llama3.2-11b-vision-instruct
lambda_ai
 131.1K $0.0150 / 1M tokens $0.0250 / 1M tokens $0.34
llama3.2-3b-instruct
lambda_ai
 131.1K $0.0150 / 1M tokens $0.0250 / 1M tokens $0.34
Llama-3.2-3B-Instruct
deepinfra
 131.1K $0.0200 / 1M tokens $0.0200 / 1M tokens $0.35
paddleocr-vl
novita
 16.4K $0.0200 / 1M tokens $0.0200 / 1M tokens $0.35
Meta-Llama-3.1-8B-Instruct-Turbo
deepinfra
 131.1K $0.0200 / 1M tokens $0.0300 / 1M tokens $0.43

nscale

Qwen2.5-Coder-3B-Instruct

In this workload, the estimated monthly API cost is $0.33. The route's listed context window is N/A.

Open model details

nscale

Qwen2.5-Coder-7B-Instruct

In this workload, the estimated monthly API cost is $0.33. The route's listed context window is N/A.

Open model details

nebius

Qwen2.5-Coder-7B

In this workload, the estimated monthly API cost is $0.33. The route's listed context window is 32.8K.

Open model details

Related workloads

Compare adjacent workload guides

Coding tasks are context-heavy and output-heavy. These guides show how the model shortlist changes when input shrinks, context grows, or document length dominates.

Chatbot

500-token chatbot workload

Short-context, high-volume support or product chat cost screen.

RAG

2,100-token RAG context

Compare retrieval-heavy answer costs when context windows dominate the bill.

Coding agent

7k-input coding-agent workload

A narrower multi-turn coding-agent scenario with repo context and task volume.

Change the workload before choosing a route

If your coding tasks use longer prompts, more iterations, or fewer monthly users, rerun the calculator with your own token counts.

Open coding-agent calculator Compare top three

Caveats

What this comparison does not prove

This page does not rank code-generation quality, latency, tool calling, IDE integration, repository understanding, or rate limits. Some low-cost routes may be specialized, gated, or inappropriate for production coding workflows. Use this as a pricing shortlist, then test the exact model route and verify final pricing with the provider.