Menu

Guides · dev · minimax api pricing

MiniMax M3: the OpenRouter price, and what it is not

Published 16 Oct 2026. Updated 16 Oct 2026. Researched 16 Oct 2026.

MiniMax M3 is the text model people mean when they say MiniMax got cheap. The number has to come from the snapshot, because we did not save a second price from minimax.io’s billing page. If those two diverge the way GLM-5.3 does, this page should grow an endpoints note. It does not have one yet.

MiniMax text ids on OpenRouterminimax-m3$0.525/Mminimax-m2.7$0.525/Mminimax-m2.5$0.440/Mclaude-opus-5.5$8.00/Mgpt-6.1-sol$4.00/M
MiniMax text ids on OpenRouter. OpenRouter list price, blended 75% input / 25% output. Snapshot 2026-10-16 05:20 UTC. Comparators resolve to Anthropic: Claude Opus 5.5 and OpenAI: GPT-6.1 Sol.
ModelBlendedvs Opusvs Sol
minimax-m3$0.52593% cheaper87% cheaper
minimax-m2.7$0.52593% cheaper87% cheaper
minimax-m2.5$0.44095% cheaper89% cheaper
Illustrative Coding-agent session costminimax-m3$0.324Anthropic: Claude Opus 5.5$2.94OpenAI: GPT-6.1 Sol$1.47
Coding-agent session: 3 million input tokens, 90% served from cache, 60 thousand output tokens. Cache-write fees are not included. All rows use the OpenRouter list price. Comparators are OpenRouter list prices for Anthropic: Claude Opus 5.5 ($8.00/M blended) and OpenAI: GPT-6.1 Sol. Snapshot 2026-10-16. Cache-write fees are excluded. A missing cache price is billed as a full input token.
ModelSessionvs Opusvs SolCache price
minimax-m3$0.32489% cheaper78% cheaperpublished
Anthropic: Claude Opus 5.5$2.94——published
OpenAI: GPT-6.1 Sol$1.47——published

The percentages in this table compare session cost. The price table's percentages compare blended token price. They answer different questions.

Where it sits in public evals

On the LMArena style-control snapshot, minimax-m3 is present in both the overall and coding categories. It is not in the top cluster with Kimi K3. It is a mid-pack preference score at a flash-adjacent price. That can be a good trade. It is not “as preferred as the latest Opus”.

Coding category, including MiniMax M3Claude Opus 5.5 (High)1547.52gpt-6.1-sol-max1539.79Kimi K3 (Max)1537.83mimo-v2.6-pro1537.58glm-5.3-flash1524.53deepseek-v4.1-flash-max1523.37qwen3.8-max1523.36glm-5.3-max1518.51kimi-k2.61516.03mimo-v2.6-flash1512.93deepseek-v4-pro1500.58qwen3.8-27b1499.13minimax-m31496.69deepseek-v4-flash1484.16hunyuan-hy3-preview1452.74
Style-controlled Bradley-Terry ratings from the LMArena leaderboard dataset. Not Elo. A higher rating means the model was preferred more often in blind votes. Intervals are the dataset's rating_lower and rating_upper. Ratings are rounded to 1 decimal for display in the chart; full values are stored. Category: coding. Published 2026-10-15, fetched 2026-10-16. Source: https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, latest).
EntryratingDetail
Claude Opus 5.5 (High)1547.52rank 6, 1,894 votes, 1533.61–1561.44
gpt-6.1-sol-max1539.79rank 10, 1,257 votes, 1522.63–1556.95
Kimi K3 (Max)1537.83rank 12, 8,306 votes, 1530.58–1545.09
mimo-v2.6-pro1537.58rank 13, 1,105 votes, 1519.93–1555.23
glm-5.3-flash1524.53rank 27, 8,457 votes, 1517.07–1531.99
deepseek-v4.1-flash-max1523.37rank 29, 3,527 votes, 1513.1–1533.65
qwen3.8-max1523.36rank 30, 7,677 votes, 1515.79–1530.93
glm-5.3-max1518.51rank 38, 5,400 votes, 1509.88–1527.13
kimi-k2.61516.03rank 44, 11,091 votes, 1509.15–1522.92
mimo-v2.6-flash1512.93rank 51, 2,625 votes, 1500.97–1524.89
deepseek-v4-pro1500.58rank 78, 17,136 votes, 1494.66–1506.51
qwen3.8-27b1499.13rank 81, 6,417 votes, 1491.03–1507.23
minimax-m31496.69rank 87, 18,242 votes, 1490.99–1502.39
deepseek-v4-flash1484.16rank 105, 15,288 votes, 1477.95–1490.37
hunyuan-hy3-preview1452.74rank 153, 2,055 votes, 1439.33–1466.16

SWE-bench Verified in our extract has MiniMax M2.5 at 75.8 resolved on 17 Feb 2026, in a mini-agent submission, next to Claude rows from the same date. There is no M3 row in the names we kept, and we did not see one in the Verified name scan. Do not promote M2.5’s score into an M3 claim.

What it is good at, on this evidence

Good candidate for bulk coding help and drafts when you want a named lab and a low blended price, and you have already rejected the mystery cheapest host on a flagship id. Weak candidate if you need a published SWE-bench M3 number or a video model. For video, read Hailuo and ignore this chart.

We did not confirm card payment or a non-China processing region on MiniMax’s own console. OpenRouter is the path this price refers to.

Related: which model, new entrants.

Questions

How much is MiniMax M3?

The figure on this page is the OpenRouter list price from the build snapshot. We did not archive a separate MiniMax invoice page for M3 on 16 Oct 2026.

Is M3 the same as Hailuo?

No. Hailuo is the video product. M3 is a text model.

Sources

  1. OpenRouter models API, accessed 2026-10-16
  2. MiniMax, accessed 2026-10-16
  3. LMArena leaderboard dataset, accessed 2026-10-16
  4. SWE-bench leaderboards.json, accessed 2026-10-16