DeepInfra raises $107M Series B to scale the inference cloud — read the announcement

Updated November 2026.
From 19.0 to 71.9 on DeepSWE — that’s the jump Xiaomi reports between MiMo-V2.5-Pro and MiMo-V2.6-Pro, its new flagship, on real-world software engineering tasks. That puts it, on Xiaomi’s own published figures, alongside Claude Opus 5 (74.0) and GPT-5.6 Sol (73.0) on a benchmark designed to test repository-level coding rather than toy problems. Built by Xiaomi’s MiMo team, it’s a sparse Mixture-of-Experts model with 1.02T total parameters and 42B activated per forward pass, designed for long-horizon agentic workloads across a 1-million-token context window. It handles text, image, video, and audio natively — no bolt-on adapters.
The training approach is what makes the architecture choices pay off. Xiaomi calls it “You Only RL Once”: a single RL run that mixes coding, general agent, visual, and cybersecurity tasks in the same batch rather than running separate per-domain phases — so strategies transfer across domains within one training run. That shows up in Xiaomi’s own reported numbers: AutomationBench goes from 16.0 to 53.1, CyberGym from 40.0 to 94.0, versus the previous version. It ships under an MIT license, which removes the licensing friction that typically complicates production deployments at this parameter scale. MiMo-V2.6-Pro is now available on DeepInfra.
MiMo-V2.6-Pro is a sparse Mixture-of-Experts model with 1.02T total parameters, but only 42B are activated per forward pass. The LLM backbone runs 70 layers (60 sliding-window attention / 10 global attention), with 384 routed experts and 8 activated per token — no shared experts. It’s natively omnimodal, processing text, image, video, and audio through dedicated encoders: a 681M-parameter MiMo ViT for vision and two audio encoders (an AudioTokenizer at 308M params and an audio patch encoder at 127M). The context window is 1,048,576 tokens (1M), which matters practically for long repository ingestion, multi-tool agent traces, and multi-session runs.
The training story is worth understanding. Xiaomi calls it “You Only RL Once”: a single mixed RL run across coding, general agent tasks, visual, and cybersecurity domains simultaneously — not separate per-domain fine-tuning passes. Tasks are mixed in the same batch, so the model can transfer reasoning strategies across domains. This is complemented by Groupwise Agentic Grading, a self-improvement loop that builds task-specific rubrics from contrasting rollouts and re-ranks passing trajectories to steer toward shorter, more token-efficient solutions. A speculative decoder (5-layer SWA MTP drafter) predicts 7 subsequent tokens per forward pass, improving throughput without changing output quality. For teams running RL on their own agents, our multi-turn reinforcement learning guide explains why rollouts are mostly an inference problem.
The improvement over MiMo-V2.5-Pro is substantial across the board, per Xiaomi’s own benchmark comparisons:
| Benchmark | V2.5 Pro | V2.6 Pro | Δ |
|---|---|---|---|
| DeepSWE v1.1 | 19.0 | 71.9 | +52.9 |
| AutomationBench v1.0.6 | 16.0 | 53.1 | +37.1 |
| Terminal Bench 4.0 | 1.5 | 34.9 | +33.4 |
| JobBench | 25.0 | 62.0 | +37.0 |
| CyberGym | 40.0 | 94.0 | +54.0 |
| MiMo Cyber Bench | 0.0 | 80.2 | +80.2 |
Those aren’t incremental gains — the cybersecurity and agentic terminal benchmarks in particular suggest a substantial capability shift rather than tuning on familiar evals.
Against frontier competitors, V2.6-Pro is competitive on agentic tasks while trailing on others. On Xiaomi’s published benchmark comparisons, it beats Claude Opus 5 (50.3) and GPT-5.6 Sol (45.8) on AutomationBench with a score of 53.1. On MiMo VisualCoding (72.3) it edges out Claude Opus 5 (70.0). It falls behind on ExploitBench (47.9 vs. 70–78.5 for Claude Opus 5 and GPT-5.6 Sol) and Terminal Bench 4.0 (34.9 vs. 39.9–49.0). The model’s strength is clearly in multi-step agentic workflows rather than single-turn reasoning benchmarks.
Xiaomi also introduced MiMo-V2.6-Pro-UltraSpeed, a serving mode on its own API built for substantially faster output than standard Pro. It’s not a separate model; it’s the same Pro checkpoint served differently, designed for real-time and latency-sensitive workflows, and Xiaomi offers it at a higher cost per token through its API and MiMo Desktop. DeepInfra serves the standard MiMo-V2.6-Pro.
MiMo-V2.6-Pro is available now on DeepInfra. With the current 10% promotion, pricing is $0.387 per million input tokens (cache miss), $0.00324 per million cached input tokens, and $0.783 per million output tokens; list prices are $0.43, $0.0036 and $0.87 (pricing as of 2026-11-17).
Need a cheaper model for simple agent steps? MiMo-V2.6-Flash is on DeepInfra at $0.14 / $0.28 per million input / output tokens, about 64% less than Pro at current prices.
If your workload involves long, repetitive context — agent traces, large codebases, multi-turn sessions — the prompt cache makes a real difference at that input price.
DeepInfra exposes the model through an OpenAI-compatible API with usage-based billing and no infrastructure to manage. It plugs in wherever you already use the OpenAI SDK. Private endpoint deployment is also available through the DeepInfra dashboard for workloads that need dedicated capacity. DeepInfra operates with a zero-retention policy on all inference requests and is SOC 2 and ISO 27001 certified.
Here’s a minimal example to get your first call running. Export your DeepInfra API key as DEEPINFRA_API_KEY and you’re set:
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["DEEPINFRA_API_KEY"],
base_url="https://api.deepinfra.com/v1/openai",
)
response = client.chat.completions.create(
model="XiaomiMiMo/MiMo-V2.6-Pro",
messages=[{"role": "user", "content": "Outline a research plan for synthesizing a MOF optimized for PFAS adsorption."}],
temperature=1.0,
top_p=0.95,
)
print(response.choices[0].message.content)Shell
curl "https://api.deepinfra.com/v1/openai/chat/completions" \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $DEEPINFRA_API_KEY" \
-d '{
"model": "XiaomiMiMo/MiMo-V2.6-Pro",
"messages": [
{
"role": "user",
"content": "Outline a research plan for synthesizing a MOF optimized for PFAS adsorption."
}
],
"temperature": 1.0,
"top_p": 0.95
}'JavaScript
import OpenAI from "openai";
const openai = new OpenAI({
apiKey: process.env.DEEPINFRA_API_KEY,
baseURL: "https://api.deepinfra.com/v1/openai",
});
const response = await openai.chat.completions.create({
model: "XiaomiMiMo/MiMo-V2.6-Pro",
messages: [{ role: "user", content: "Outline a research plan for synthesizing a MOF optimized for PFAS adsorption." }],
temperature: 1.0,
top_p: 0.95,
});
console.log(response.choices[0].message.content);Xiaomi’s own deployment guidance recommends temperature=1.0 and top_p=0.95 — worth sticking to those defaults, especially for agentic and reasoning workloads where the model benefits from the full output distribution. Head to the MiMo-V2.6-Pro page on DeepInfra to grab your API key and start building.
MiMo-V2.6-Pro is a serious agentic model — the benchmark jumps reflect genuine capability changes rather than eval overfitting, and the MIT license means you can ship it without legal back-and-forth. The combination of a 1M-token context window, native function calling, and training that explicitly optimizes for multi-step tool use makes it a reasonable fit for the class of problems that single-turn models still struggle with: long-running coding agents, automated security workflows, and complex multi-session tasks that require coherent reasoning across large contexts.
If you’re building anything in the agentic space, this is worth running through your eval suite — and worth comparing against other recent releases like DeepSeek V4 Flash and Qwen3.6 or the Kimi K3 vs. Claude Opus 4.8 vs. GPT-5.6 Sol lineup. Start at the MiMo-V2.6-Pro model page on DeepInfra, or browse the full model catalog for other options.
GLM-5 API Benchmarks: Latency, Throughput & Cost<p>Updated November 2026. GLM-5 has been retired on DeepInfra. Its successor, GLM-5.3, is available now; see the GLM-5.3 API provider benchmarks for current figures. The benchmark analysis below is kept for reference. GLM-5 is the latest open-weights reasoning model released by Z AI (Zhipu AI) in February 2026, characterized by high “thinking token” usage. It […]</p>
FLUX.1-dev Guide: Mastering Text-to-Image AI Prompts for Stunning and Consistent VisualsLearn how to craft compelling prompts for FLUX.1-dev to create stunning images.© 2026 DeepInfra. All rights reserved.