Cheapest AI model for coding agents

An agentic loop that reads files and writes patches, generating heavy output every turn.

List prices checked

How this is scored

Scored at 400M input and 100M output tokens per month. Reasoning and generated code both bill as output, so this ranking is driven by the output rate — models that look cheap on input alone fall a long way down this table.

All 19, ranked for this workload

#ModelProviderCost for this workloadNotes
1DeepSeek-V4-FlashDeepSeek$84/moA million-token context window at a fraction of any Western model's price, with cache-hit input billed at a fiftieth of the base rate. DeepSeek has published notice of a significant price rise, so treat these rates as a floor rather than a plan.
2Grok 3 minixAI$170/moAn unusually cheap reasoning model, with output priced below most competitors' input rates — which matters because reasoning models are output-heavy by nature.
3CodestralMistral AI$210/moA code-specialised model tuned for fill-in-the-middle completion, priced for the high request volume that inline editor autocomplete generates.
4DeepSeek-V4-ProDeepSeek$261/moFrontier-class reasoning at roughly a fifth of Western frontier pricing, with a million-token window. The trade-offs are Chinese data residency, a 500-request concurrency cap, and a price rise the vendor has already announced.
5GPT-5 miniOpenAI$300/moA fifth of GPT-5's price with the same 400k context window, which makes it the cheapest way to process very long documents without dropping to a small model.
6Gemini 2.5 FlashGoogle$370/moThe best price-to-context ratio on the market: a million tokens of multimodal input for a fraction of what frontier models charge, with optional thinking budgets.
7OpenAI o4-miniOpenAI$880/moCheap reasoning, strong on maths and code relative to its price. Same caveat as every reasoning model: budget for the thinking tokens, not just the visible answer.
8Claude Haiku 4.5Anthropic$900/moThe fast, cheap end of the Claude family, priced at a third of Sonnet while keeping the same context window. Suited to classification, extraction, and high-volume routing.
9Mistral LargeMistral AI$1,400/moThe French lab's flagship, priced below every US frontier model and hosted in the EU, which is often the deciding factor rather than the benchmark scores.
10GPT-5OpenAI$1,500/moOpenAI's flagship, priced aggressively below the other frontier models on input and with a 90% cached-input discount that makes long system prompts nearly free to reuse.
11Gemini 2.5 ProGoogle$1,500/moA million-token context window with native video and audio input, priced at frontier-model rates below 200k tokens and at a higher tier above it — a detail that surprises people modelling long-context costs.
12GPT-4.1OpenAI$1,600/moA million-token context window at a mid-tier price, which remains the reason to reach for it over newer models when the whole codebase has to fit in one request.
13OpenAI o3OpenAI$1,600/moA reasoning model that spends tokens thinking before answering. The list price understates the real cost, because reasoning tokens bill as output and can dominate the total.
14GPT-4oOpenAI$2,000/moThe natively multimodal model that set the mid-market price for over a year. Now more expensive than newer models that outperform it, but still the default in a great deal of shipped code.
15Cohere Command ACohere$2,000/moCohere's enterprise flagship, built around retrieval-augmented generation and multilingual work, and sold heavily as a private deployment inside customer infrastructure.
16Claude Sonnet 5Anthropic$2,700/moThe workhorse of the Claude family and the default for production work at volume. Introductory pricing of $2/$10 per million tokens runs to 31 August 2026; the $3/$15 standard rate shown here is what applies after that.
17Grok 4xAI$2,700/moxAI's flagship, priced to match Claude Sonnet exactly, with a large context window and live access to posts on X as a differentiator.
18Claude Opus 5Anthropic$4,500/moAnthropic's model for complex agentic coding and enterprise work. Prompt caching cuts repeated input to a tenth of the base rate, which for an agent resending the same context every turn matters more than the headline price.
19Claude Fable 5Anthropic$9,000/moAnthropic's top model for long-running agentic work, priced at twice Opus 5. Worth it only where a longer autonomous run genuinely replaces human supervision, since the token cost compounds across a multi-hour session.

Published list prices only. Rows that do not publish a rate for an axis this workload depends on are excluded rather than shown as cheap.

Questions

Cheapest AI model for coding agents in 2026?
DeepSeek-V4-Flash from DeepSeek — $84/mo. Grok 3 mini is second at $170/mo. This ranking is specific to the workload described above; a different usage shape reorders it.
How is this ranking weighted?
Scored at 400M input and 100M output tokens per month. Reasoning and generated code both bill as output, so this ranking is driven by the output rate — models that look cheap on input alone fall a long way down this table.
Why does this differ from the general cheapest list?
Because a single price axis never describes a real workload. Ranking by headline rate answers "who is cheapest per unit"; this page answers "who is cheapest for this job", and the two orders are often very different — which is the whole reason it exists as a separate table.
What is not accounted for?
Committed-use discounts, regional price variation, per-request charges, support plans, and minimum retention. These are published list prices applied to one stated workload — a shortlist to price properly with the vendor, not a quotation.

Other workloads