Budget
US$0.20
Qwen2.5 Coder 32B
Qwen (Alibaba)
Meta (via aggregator) · Budget · US$0.88 input/1M · US$0.88 output/1M · 128K context
Typical monthly cost
US$18.26
≈ per day
US$0.83
Blended cost/1M
US$0.88
Context window
128K
Llama 3.3 70B Instruct from Meta (via aggregator) is a budget model priced at US$0.88 per 1M input tokens and US$0.88 per 1M output tokens. For a typical solo-developer workload (8 hours/day, 22 days/month — 1 medium feature, 5 small bug fixes, 4 PR reviews, 2 stack-trace debugs, ~1500 lines of TypeScript, 1 large-doc read, with prompt caching at the default mix) Llama 3.3 70B Instruct costs about US$18/month. The 128K-token context window covers single-file workloads comfortably.
Move the slider or switch task mix — values update live.
Monthly budget
US$100 / month
≈ $23/wk · ≈ $4.55/day
on Llama 3.3 70B Instruct
Input tokens
96.6M
Output tokens
17.0M
Total tokens
113.6M
Per month this budget delivers
At the default coding-agent mix.
The 22-day month is based on the median working-day count across DE/US.
| Activity | Count | Per task | Daily | Monthly |
|---|---|---|---|---|
| Medium feature (10–15 files) | 1 | US$0.24 | US$0.24 | US$5.23 |
| Small bug fix | 5 | US$0.03 | US$0.14 | US$3.02 |
| PR review | 4 | US$0.03 | US$0.11 | US$2.48 |
| Debug from stack trace | 2 | US$0.06 | US$0.12 | US$2.59 |
| Read a large doc | 1 | US$0.05 | US$0.05 | US$1.00 |
| Micro-interaction (explain / lint fix) | 30 | US$0.00 | US$0.03 | US$0.70 |
| Lines of TypeScript | 1,500 | US$0.00 | US$0.15 | US$3.25 |
| Total | US$0.83 | US$18.26 | ||
The 1500-lines-of-TS row models ~1000 lines read (cache-hit) + ~500 lines written. Headline figures are precise to ~5% — see the FAQ.
What each monthly budget buys on this model (typical solo-developer day, 22 working days).
| Monthly budget | Medium features | PR reviews | Debug sessions | Lines of TS |
|---|---|---|---|---|
| Typical (≈ $18) | 76 | 648 | 310 | 185,310 |
| $50/month | 210 | 1,775 | 848 | 507,305 |
| $200/month | 841 | 7,102 | 3,394 | 2,029,220 |
| $500/month | 2,104 | 17,755 | 8,486 | 5,073,051 |
| $2000/month | 8,417 | 71,022 | 33,946 | 20,292,207 |
Typical mix: coding-agent (85% input, 50% cache hits). Values show the maximum count of each task type at that budget.
Coding
Trained or post-trained for code generation tasks.
Reasoning
Not supported
Multimodal
Not supported
Prompt cache
Not supported
Batch API
Not supported
Tool use
Native function-calling / tool-use API support.
Long context
Not supported
Extended thinking
Not supported
Total models
1
Median input/1M
US$0.88
Median output/1M
US$0.88
Input range
US$0.88–US$0.88
Verified: 2026-05-07
Running the realistic solo-developer day (1 medium feature + 5 small bug fixes + 4 PR reviews + 2 debug sessions + ~1500 lines of TypeScript + 1 large-doc read, 22 working days) on Llama 3.3 70B Instruct costs about US$18/month. Heavier workloads scale proportionally; lighter workloads cost less.
128K tokens total, with up to 8K of output. That fits a few dozen source files in a single call.
Providers charge US$0.88 per 1M output tokens against US$0.88 per 1M input — output requires real compute, input comes mostly from cache. Coding agents read many files (input-heavy) and emit compact diffs (low output), so total spend is usually input-driven.
No. Every input token is billed at the full US$0.88/1M rate. If your workload reuses the same system prompt frequently, compare against a caching-capable model (Claude Sonnet 4.6, GPT-5) where the effective input rate falls sharply.
Open the full calculator with your own budget, task mix and region (US or DE with 19% VAT).
Open calculator