Engineering
A 500-line diff with surrounding file context in; review comments out.
Pricing verified:
Between $0.0005 and $0.082 per run, depending on the model. That is a 164× spread for the same job.
A 500-line diff plus enough surrounding context to review it honestly ≈ 12,000 input tokens. Review comments on a change that size ≈ 900 output tokens.
Cost = (input tokens ÷ 1M × input price) + (output tokens ÷ 1M × output price). Prices come from our daily-verified model data. Batch and cached-input discounts are not applied — they only ever make these numbers smaller.
| Model | Provider | Per run | At 400 PRs a month | Context |
|---|---|---|---|---|
| Gemma 2 9Bneeds chunking | $0.0004 | $0.176 | 8k tokens | |
| Llama 3.2 1B Instruct | Meta | $0.0005 | $0.202 | 60k tokens |
| Llama 3.1 8B Instruct | Meta | $0.0007 | $0.269 | 16k tokens |
| GPT-5 Nano | OpenAI | $0.0010 | $0.384 | 400k tokens |
| gpt-oss-safeguard-20b | OpenAI | $0.0010 | $0.408 | 131k tokens |
| Mistral Small 3.2 24B | Mistral | $0.0011 | $0.432 | 128k tokens |
| Gemini 2.0 Flash Lite | $0.0012 | $0.468 | 1.0M tokens | |
| Ministral 3 3B 2512 | Mistral | $0.0013 | $0.516 | 131k tokens |
| Devstral Small 1.1 | Mistral | $0.0015 | $0.588 | 131k tokens |
| Mistral Small Creative | Mistral | $0.0015 | $0.588 | 33k tokens |
| Voxtral Small 24B 2507 | Mistral | $0.0015 | $0.588 | 32k tokens |
| Mistral Small 3.1 | Mistral | $0.0015 | $0.588 | 128k tokens |
| Mistral 7B Instruct v0.1needs chunking | Mistral | $0.0015 | $0.596 | 3k tokens |
| Gemini 2.0 Flash | $0.0016 | $0.624 | 1.0M tokens | |
| Gemini 2.5 Flash Lite | $0.0016 | $0.624 | 1.0M tokens | |
| Gemini 2.5 Flash Lite Preview 09-2025 | $0.0016 | $0.624 | 1.0M tokens | |
| GPT-4.1 Nano | OpenAI | $0.0016 | $0.624 | 1.0M tokens |
| Llama 3 8B Instructneeds chunking | Meta | $0.0018 | $0.722 | 8k tokens |
| Gemma 4 26B A4B | $0.0019 | $0.768 | 262k tokens | |
| DeepSeek V4-Flash | DeepSeek | $0.0019 | $0.773 | 1M tokens |
| Ministral 3 8B 2512 | Mistral | $0.0019 | $0.774 | 262k tokens |
| Mistral Nemo | Mistral | $0.0019 | $0.774 | 131k tokens |
| GLM-5.3 Flash | Z.ai | $0.0022 | $0.900 | 1M tokens |
| Llama Guard 4 12B | Meta | $0.0023 | $0.929 | 164k tokens |
| GPT-4o Mini | OpenAI | $0.0023 | $0.936 | 128k tokens |
| Mistral Small 4 | Mistral | $0.0023 | $0.936 | 262k tokens |
| Qwen 3.8 Flash | Alibaba | $0.0023 | $0.937 | 991k tokens |
| Ministral 3 14B 2512 | Mistral | $0.0026 | $1.03 | 262k tokens |
| Saba | Mistral | $0.0029 | $1.18 | 33k tokens |
| GPT-5.6 Luna | OpenAI | $0.0035 | $1.39 | 1.1M tokens |
| Grok Code Fast 1 | xAI | $0.0037 | $1.50 | 256k tokens |
| Grok 3 Mini | xAI | $0.0040 | $1.62 | 131k tokens |
| Grok 3 Mini Beta | xAI | $0.0040 | $1.62 | 131k tokens |
| Claude 3 Haiku | Anthropic | $0.0041 | $1.65 | 200k tokens |
| DeepSeek V3 | DeepSeek | $0.0042 | $1.69 | 128k tokens |
| Gemini 3 Flash Preview | $0.0043 | $1.74 | 1.0M tokens | |
| Codestral 2508 | Mistral | $0.0044 | $1.76 | 256k tokens |
| Llama 3.2 11B Vision Instruct | Meta | $0.0045 | $1.78 | 131k tokens |
| Mistral Small 3 | Mistral | $0.0047 | $1.88 | 33k tokens |
| GPT-5 Mini | OpenAI | $0.0048 | $1.92 | 400k tokens |
| GPT-5.1-Codex-Mini | OpenAI | $0.0048 | $1.92 | 400k tokens |
| Llama 3.1 70B Instruct | Meta | $0.0052 | $2.06 | 131k tokens |
| Muse Glimmer 30B | Meta | $0.0055 | $2.22 | 131k tokens |
| Gemma 4 31B | $0.0056 | $2.22 | 262k tokens | |
| Llama Guard 3 8B | Meta | $0.0058 | $2.31 | 131k tokens |
| Gemini 2.5 Flash | $0.0058 | $2.34 | 1.0M tokens | |
| Gemini 3.5 Flash-Lite | $0.0058 | $2.34 | 1.0M tokens | |
| Nano Banana (Gemini 2.5 Flash Image) | $0.0058 | $2.34 | 33k tokens | |
| DeepSeek V4-Pro | DeepSeek | $0.0060 | $2.40 | 1M tokens |
| GPT-4.1 Mini | OpenAI | $0.0062 | $2.50 | 1.0M tokens |
| Devstral 2 2512 | Mistral | $0.0066 | $2.64 | 262k tokens |
| Devstral Medium | Mistral | $0.0066 | $2.64 | 131k tokens |
| Mistral Medium 3 | Mistral | $0.0066 | $2.64 | 131k tokens |
| Mistral Medium 3.1 | Mistral | $0.0066 | $2.64 | 131k tokens |
| Llama 3 70B Instructneeds chunking | Meta | $0.0068 | $2.71 | 8k tokens |
| Mixtral 8x7B Instruct | Mistral | $0.0070 | $2.79 | 33k tokens |
| Llama 4 Scout | Meta | $0.0071 | $2.83 | 512k tokens |
| GPT-3.5 Turbo | OpenAI | $0.0074 | $2.94 | 16k tokens |
| Mistral Large 3 2512 | Mistral | $0.0074 | $2.94 | 262k tokens |
| Gemma 2 27Bneeds chunking | $0.0084 | $3.35 | 8k tokens | |
| DeepSeek R1 | DeepSeek | $0.0086 | $3.43 | 128k tokens |
| Llama 4 Maverick | Meta | $0.0086 | $3.46 | 256k tokens |
| Gemini 3.1 Flash | $0.0087 | $3.48 | 1M tokens | |
| GPT Audio Mini | OpenAI | $0.0094 | $3.74 | 128k tokens |
| Gemini 3.6 Flash | $0.012 | $4.95 | 1.0M tokens | |
| Gemini 3.7 Flash | $0.012 | $4.95 | 1.0M tokens | |
| Claude 3.5 Haiku | Anthropic | $0.013 | $5.28 | 200k tokens |
| Claude 4 Haiku | Anthropic | $0.013 | $5.28 | 200k tokens |
| Codestral 25.01 | Mistral | $0.013 | $5.29 | 256k tokens |
| GPT-3.5 Turbo (older v0613)needs chunking | OpenAI | $0.014 | $5.52 | 4k tokens |
| Kimi K2.7 Code | Moonshot | $0.015 | $6.00 | 256k tokens |
| Claude Haiku 4.5 | Anthropic | $0.017 | $6.60 | 200k tokens |
| o3 Mini | OpenAI | $0.017 | $6.86 | 200k tokens |
| o3 Mini High | OpenAI | $0.017 | $6.86 | 200k tokens |
| o4 Mini | OpenAI | $0.017 | $6.86 | 200k tokens |
| o4 Mini High | OpenAI | $0.017 | $6.86 | 200k tokens |
| GPT-5.2 Mini | OpenAI | $0.019 | $7.49 | 128k tokens |
| Muse Spark | Meta | $0.019 | $7.53 | 1.0M tokens |
| GPT-3.5 Turbo Instructneeds chunking | OpenAI | $0.020 | $7.92 | 4k tokens |
| GLM-5.2 | Z.ai | $0.021 | $8.30 | 1M tokens |
| GLM-5.3 | Z.ai | $0.021 | $8.30 | 1M tokens |
| GPT-5 | OpenAI | $0.024 | $9.60 | 400k tokens |
| GPT-5 Chat | OpenAI | $0.024 | $9.60 | 128k tokens |
| GPT-5 Codex | OpenAI | $0.024 | $9.60 | 400k tokens |
| GPT-5.1 | OpenAI | $0.024 | $9.60 | 400k tokens |
| GPT-5.1 Chat | OpenAI | $0.024 | $9.60 | 128k tokens |
| GPT-5.1-Codex | OpenAI | $0.024 | $9.60 | 400k tokens |
| GPT-5.1-Codex-Max | OpenAI | $0.024 | $9.60 | 400k tokens |
| Gemini 2.5 Pro | $0.024 | $9.60 | 1.0M tokens | |
| Gemini 2.5 Pro Preview 05-06 | $0.024 | $9.60 | 1.0M tokens | |
| Gemini 2.5 Pro Preview 06-05 | $0.024 | $9.60 | 1.0M tokens | |
| Mistral Medium 3.5 | Mistral | $0.025 | $9.90 | 256k tokens |
| Gemini 3.5 Flash | $0.026 | $10.44 | 1.0M tokens | |
| Grok 4 | xAI | $0.029 | $11.76 | 2M tokens |
| Grok 4.5 | xAI | $0.029 | $11.76 | 500k tokens |
| Grok 4.6 | xAI | $0.029 | $11.76 | 500k tokens |
| Mixtral 8x22B Instruct | Mistral | $0.029 | $11.76 | 66k tokens |
| Pixtral Large 2411 | Mistral | $0.029 | $11.76 | 131k tokens |
| Qwen 3.8 Max | Alibaba | $0.029 | $11.76 | 1M tokens |
| GPT-4.1 | OpenAI | $0.031 | $12.48 | 1.0M tokens |
| o3 | OpenAI | $0.031 | $12.48 | 200k tokens |
| o4 Mini Deep Research | OpenAI | $0.031 | $12.48 | 200k tokens |
| GPT-5 Image Mini | OpenAI | $0.032 | $12.72 | 400k tokens |
| Claude Sonnet 5 | Anthropic | $0.033 | $13.20 | 1M tokens |
| GPT-5.6 Sol | OpenAI | $0.033 | $13.20 | 1.1M tokens |
| GPT-5.2 | OpenAI | $0.034 | $13.44 | 200k tokens |
| GPT-5.3 Chat | OpenAI | $0.034 | $13.44 | 128k tokens |
| GPT-5.3-Codex | OpenAI | $0.034 | $13.44 | 400k tokens |
| GPT-5.6 Terra | OpenAI | $0.035 | $13.92 | 1.1M tokens |
| Gemini 3.1 Pro | $0.035 | $13.92 | 2M tokens | |
| Nano Banana Pro (Gemini 3 Pro Image Preview) | $0.035 | $13.92 | 66k tokens | |
| Qwen 3.7 Max | Alibaba | $0.037 | $14.70 | 1M tokens |
| GPT Audio | OpenAI | $0.039 | $15.60 | 128k tokens |
| GPT-4o | OpenAI | $0.039 | $15.60 | 128k tokens |
| GPT-3.5 Turbo 16k | OpenAI | $0.040 | $15.84 | 16k tokens |
| GPT-5.4 | OpenAI | $0.043 | $17.40 | 272k tokens |
| Mistral Large 2 | Mistral | $0.044 | $17.64 | 128k tokens |
| Claude 3.7 Sonnet (thinking) | Anthropic | $0.050 | $19.80 | 200k tokens |
| Claude Sonnet 4 | Anthropic | $0.050 | $19.80 | 200k tokens |
| Claude Sonnet 4.5 | Anthropic | $0.050 | $19.80 | 1M tokens |
| Claude Sonnet 4.6 | Anthropic | $0.050 | $19.80 | 1M tokens |
| Grok 3 | xAI | $0.050 | $19.80 | 131k tokens |
| Grok 3 Beta | xAI | $0.050 | $19.80 | 131k tokens |
| Kimi K3 | Moonshot | $0.050 | $19.80 | 1M tokens |
| Claude Opus 4.5 | Anthropic | $0.082 | $33.00 | 200k tokens |
| Claude Opus 4.6 | Anthropic | $0.082 | $33.00 | 1M tokens |
| Claude Opus 4.7 | Anthropic | $0.082 | $33.00 | 1M tokens |
| Claude Opus 4.8 | Anthropic | $0.082 | $33.00 | 1M tokens |
| Claude Opus 5 | Anthropic | $0.082 | $33.00 | 1M tokens |
| GPT-5.5 | OpenAI | $0.087 | $34.80 | 1M tokens |
| Claude 3.5 Sonnet | Anthropic | $0.099 | $39.60 | 200k tokens |
| GPT-5 Image | OpenAI | $0.129 | $51.60 | 400k tokens |
| GPT-4 Turbo | OpenAI | $0.147 | $58.80 | 128k tokens |
| GPT-4 Turbo (older v1106) | OpenAI | $0.147 | $58.80 | 128k tokens |
| GPT-4 Turbo Preview | OpenAI | $0.147 | $58.80 | 128k tokens |
| o3 Deep Research | OpenAI | $0.156 | $62.40 | 200k tokens |
| Claude Fable 5 | Anthropic | $0.165 | $66.00 | 1M tokens |
| Claude Fable 5.1 | Anthropic | $0.165 | $66.00 | 1M tokens |
| Claude Mythos 5 | Anthropic | $0.165 | $66.00 | 1M tokens |
| Claude Mythos 5.1 | Anthropic | $0.165 | $66.00 | 1M tokens |
| GPT-6 Astra | OpenAI | $0.165 | $66.00 | 1.1M tokens |
| o1 | OpenAI | $0.234 | $93.60 | 200k tokens |
| Claude Opus 4 | Anthropic | $0.247 | $99.00 | 200k tokens |
| Claude Opus 4.1 | Anthropic | $0.247 | $99.00 | 200k tokens |
| GPT-5 Pro | OpenAI | $0.288 | $115 | 400k tokens |
| o3 Pro | OpenAI | $0.312 | $125 | 200k tokens |
| GPT-4needs chunking | OpenAI | $0.414 | $166 | 8k tokens |
| GPT-4 (older v0314)needs chunking | OpenAI | $0.414 | $166 | 8k tokens |
| o1-pro | OpenAI | $2.34 | $936 | 200k tokens |
“Needs chunking” means the model's context window can't hold this job in a single pass, so the real cost is higher than the figure shown and quality usually suffers.
Cheap models miss the class of bug worth catching — the cross-file one that needs the surrounding context you just paid to send. Cost per review is small enough at every tier that paying up for the model with the better SWE-bench score is usually the right trade.
$0.0005 on Llama 3.2 1B Instruct, the cheapest capable option, rising to $0.082 on Claude Opus 4.8 at the top end. Mistral Small 3.2 24B is the value pick at $0.0011 per run. The job is priced at 12,000 input and 900 output tokens — see the working below.
A 500-line diff plus enough surrounding context to review it honestly ≈ 12,000 input tokens. Review comments on a change that size ≈ 900 output tokens.
$0.202 a month on Llama 3.2 1B Instruct, $0.432 on Mistral Small 3.2 24B, and $33.00 on Claude Opus 4.8. Batch APIs typically halve these figures for work that can wait, and prompt caching cuts the input side further when the same context is reused.
Cheap models miss the class of bug worth catching — the cross-file one that needs the surrounding context you just paid to send. Cost per review is small enough at every tier that paying up for the model with the better SWE-bench score is usually the right trade.
Yes. Every figure on this page is computed from our model pricing data, which is checked daily against each provider's official pricing page. When a provider changes a price, these numbers change with it.
Newsletter
Pricing changes, new model releases, and updated recommendations — delivered when it matters.
No spam. Useful updates only. Affiliate disclosures always clearly labeled.