Every text model on RouterPlex, cheapest first. Most “cheapest model” tables sort on input price, which flatters models that charge little to read and a lot to write. This one ranks on a fixed 100K-in / 10K-out agent turn, so input and output rates both count.
| # | Model | Provider | Input / 1M | Output / 1M | Blended / 1M | Per agent turn |
|---|---|---|---|---|---|---|
| 1 | DeepSeek V4 Flash1M | DeepSeek | $0.090 | $0.18 | $0.098 | $0.011 |
| 2 | MiMo V2.51M | Xiaomi | $0.11 | $0.28 | $0.13 | $0.014 |
| 3 | Hy3256K | Tencent Hunyuan | $0.20 | $0.80 | $0.25 | $0.028 |
| 4 | Step 3.7 Flash256K | StepFun | $0.20 | $1.15 | $0.29 | $0.032 |
| 5 | LongCat 2.01M | LongCat | $0.30 | $1.20 | $0.38 | $0.042 |
| 6 | MiniMax M2.7196K | MiniMax | $0.30 | $1.20 | $0.38 | $0.042 |
| 7 | MiniMax M31M | MiniMax | $0.30 | $1.20 | $0.38 | $0.042 |
| 8 | Qwen3.7 Plus1M | Alibaba | $0.32 | $1.28 | $0.41 | $0.045 |
| 9 | DeepSeek V4 Pro1M | DeepSeek | $0.43 | $0.87 | $0.47 | $0.052 |
| 10 | MiMo V2.5 Pro1M | Xiaomi | $0.44 | $0.87 | $0.48 | $0.053 |
| 11 | MiniMax M3 Highspeed1M | MiniMax | $0.45 | $1.80 | $0.57 | $0.063 |
| 12 | Qwen3.6 Plus1M | Alibaba | $0.50 | $3.00 | $0.73 | $0.080 |
| 13 | MiniMax M2.7 Highspeed196K | MiniMax | $0.60 | $2.40 | $0.76 | $0.084 |
| 14 | Doubao Seed 2.0 Code200K | ByteDance | $0.67 | $3.36 | $0.91 | $0.10 |
| 15 | Doubao Seed 2.0 Pro128K | ByteDance | $0.67 | $3.36 | $0.91 | $0.10 |
| 16 | Kimi K2.6256K | Moonshot | $0.95 | $4.00 | $1.23 | $0.14 |
| 17 | Kimi K2.7256K | Moonshot | $0.95 | $4.00 | $1.23 | $0.14 |
| 18 | Claude Haiku 4.5256K | Anthropic | $1.00 | $5.00 | $1.36 | $0.15 |
| 19 | GPT-5.6 Luna258K | OpenAI | $1.00 | $6.00 | $1.45 | $0.16 |
| 20 | GLM-5.1256K | Zhipu | $1.40 | $4.40 | $1.67 | $0.18 |
| 21 | GLM-5.21M | Zhipu | $1.40 | $4.40 | $1.67 | $0.18 |
| 22 | Gemini 3.5 Flash1M | $1.50 | $9.00 | $2.18 | $0.24 | |
| 23 | Qwen3.8 Max1M | Alibaba | $2.00 | $6.00 | $2.36 | $0.26 |
| 24 | Grok 4.5500K | xAI | $2.00 | $6.00 | $2.36 | $0.26 |
| 25 | Grok 4.6500K | xAI | $2.00 | $6.00 | $2.36 | $0.26 |
| 26 | Claude Sonnet 51M | Anthropic | $2.00 | $10.00 | $2.73 | $0.30 |
| 27 | Gemini 3.1 Pro1M | $2.00 | $12.00 | $2.91 | $0.32 | |
| 28 | Qwen3.7 Max1M | Alibaba | $2.50 | $7.50 | $2.95 | $0.33 |
| 29 | GPT-5.41M | OpenAI | $2.50 | $15.00 | $3.64 | $0.40 |
| 30 | GPT-5.6 Terra258K | OpenAI | $2.50 | $15.00 | $3.64 | $0.40 |
| 31 | Claude Sonnet 4.61M | Anthropic | $3.00 | $15.00 | $4.09 | $0.45 |
| 32 | Kimi K31M | Moonshot | $3.00 | $15.00 | $4.09 | $0.45 |
| 33 | Claude Opus 4.61M | Anthropic | $5.00 | $25.00 | $6.82 | $0.75 |
| 34 | Claude Opus 4.71M | Anthropic | $5.00 | $25.00 | $6.82 | $0.75 |
| 35 | Claude Opus 4.81M | Anthropic | $5.00 | $25.00 | $6.82 | $0.75 |
| 36 | Claude Opus 51M | Anthropic | $5.00 | $25.00 | $6.82 | $0.75 |
| 37 | GPT-5.5256K | OpenAI | $5.00 | $30.00 | $7.27 | $0.80 |
| 38 | GPT-5.6 Sol258K | OpenAI | $5.00 | $30.00 | $7.27 | $0.80 |
| 39 | Claude Fable 51M | Anthropic | $10.00 | $50.00 | $13.64 | $1.50 |
39 text models · USD per 1M tokens · agent turn = 100K input + 10K output · image models excluded (billed per image-output token)
What you pay to feed a model context — the number that matters for large-file reads and RAG.
What you pay for what the model writes — usually the dominant cost on reasoning and code generation.
Prices are vendor list rates pulled from the live catalog, the same numbers the API bills against — not a scraped snapshot that goes stale after the next price cut. Every model is scored on one fixed profile so the ordering is comparable, and image-generation models are excluded because they bill per image-output token and would not compare honestly against a chat turn.
Blended $/1M is the same turn expressed per million tokens, which is useful when you want one number to compare against a vendor's headline rate. Both columns move together; neither is a discount RouterPlex applies, because there is no markup to discount.
Price is only half the decision. A cheap model that needs three attempts at a task costs more than an expensive one that lands it first time, so check model benchmarks before moving production traffic to the top of this table.
On RouterPlex the cheapest model is DeepSeek V4 Flash at $0.090 per 1M input tokens and $0.18 per 1M output tokens — about $0.011 for a 100K-token-in, 10K-token-out agent turn. Prices are vendor list rates, so the same model costs the same whether you call it here or direct.
Ranking on input price alone favours models that charge little to read and a lot to write, which is the wrong way round for coding and agent work. This table ranks on a fixed 100K-in / 10K-out turn so input and output rates both count, and the same profile is applied to every model.
Across this catalog the spread is about 139× on the same agent turn: DeepSeek V4 Flash costs roughly $0.011 where Claude Fable 5 costs about $1.50. Whether that trade is worth it depends on task difficulty — a cheap model that needs three attempts is not cheap.
Not meaningfully. RouterPlex bills vendor list prices with no markup and no top-up fee, so per-token cost matches going direct to each vendor. What you save is the overhead of separate accounts, keys and minimum balances per provider.
RouterPlex applies no additional RPM or TPM cap beyond the upstream vendor's. The real caveat is capability: budget models are generally weaker at long agentic tool loops, so compare quality on the benchmarks pages before moving production traffic onto one.
One key reaches every model in this table. Top up from $5 and pay per token.