Chat & reasoning models — one API key, 40 models

Claude, GPT, Gemini, DeepSeek, Kimi, GLM, Qwen and Grok through the same OpenAI-compatible endpoint. Compare per-token prices side by side, then switch models by changing one string.

Crazyrouter is an AI API gateway that serves 40 chat and reasoning models through one OpenAI-compatible key, with per-token prices from $0.055 and native Anthropic, Gemini and Responses endpoints where the vendor has them.

40
models
$0.055
from, per 1M input
45%
max below list
Input price per 1M tokens (list vs Crazyrouter), cheapest first
ListCrazyrouterGemini 2.5 Flash-Lite$0.100 → $0.055GPT-4o mini$0.150 → $0.098Claude Haiku 5.5$0.100 → $0.100Gemini 3 Flash$0.500 → $0.165Gemini 2.5 Flash$0.300 → $0.165DeepSeek V4 Flash$0.300 → $0.230MiniMax M3$0.300 → $0.300GPT-4.1 mini$0.400 → $0.320MiMo V2.6 Pro$0.435 → $0.435Kimi K2.5$0.600 → $0.600GPT-5.4 mini$0.750 → $0.600Claude Haiku 4.5$1.00 → $0.650

27 more models in the table below

What the models look like in use

All chat & reasoning models, by price

ModelArena scoreOfficial priceSite discountRelease date
Kimi K3.1 ProNot matched$2.00 / $10.00—
Kimi K3.1 Pro HighspeedNot matched$4.00 / $20.00—
Claude Haiku 5.51424.8
Evaluation details
Evaluated model:
Claude Haiku 5.5 (High)
Confidence interval:
1417.3–1432.3
Battles:
7,890
$0.10 / $0.50—
GPT-6.1 Sol1481.3
Evaluation details
Evaluated model:
GPT-6.1 Sol (Max)
Confidence interval:
1474.3–1488.4
Battles:
9,367
$2.00 / $10.00—
Claude Sonnet 5.51477.1
Evaluation details
Evaluated model:
Claude Sonnet 5.5 (xHigh)
Confidence interval:
1470.4–1483.9
Battles:
10,033
$2.00 / $10.00—
GPT-6 Astra1475.6
Evaluation details
Evaluated model:
GPT-6 Astra (Max)
Confidence interval:
1469.7–1481.6
Battles:
15,038
$10.00 / $50.00−35%
Gemini 3.8 Flash1495.9
Evaluation details
Evaluated model:
Gemini 3.8 Flash (High)
Confidence interval:
1491.5–1500.4
Battles:
47,803
$0.75 / $3.75—
Grok 4.61455.5
Evaluation details
Evaluated model:
Grok 4.6 (High)
Confidence interval:
1450.9–1460.1
Battles:
33,824
$2.00 / $6.00−15%
Claude Opus 51491.6
Evaluation details
Evaluated model:
Claude Opus 5 (High)
Confidence interval:
1488.0–1495.1
Battles:
73,167
$5.00 / $25.00−35%
Kimi K31488.2
Evaluation details
Evaluated model:
Kimi K3 (Max)
Confidence interval:
1483.7–1492.6
Battles:
36,780
$3.00 / $15.00—
GPT-5.6 Sol1484.9
Evaluation details
Evaluated model:
GPT-5.6 Sol (xHigh)
Confidence interval:
1480.7–1489.1
Battles:
43,796
$4.00 / $20.00−35%
Claude Sonnet 51461.7
Evaluation details
Evaluated model:
Claude Sonnet 5 (High)
Confidence interval:
1457.7–1465.8
Battles:
52,952
$2.00 / $10.00−35%
Claude Fable 51503.9
Evaluation details
Evaluated model:
Claude Fable 5 (High)
Confidence interval:
1499.7–1508.1
Battles:
45,341
$10.00 / $50.00−35%
MiniMax M31439.2
Evaluation details
Evaluated model:
MiniMax-M3
Confidence interval:
1435.5–1443.0
Battles:
68,611
$0.30 / $1.20—
Claude Opus 4.81474.4
Evaluation details
Evaluated model:
Claude Opus 4.8
Confidence interval:
1470.6–1478.3
Battles:
65,121
$5.00 / $25.00−35%
GPT-5.51476.7
Evaluation details
Evaluated model:
GPT-5.5
Confidence interval:
1473.0–1480.3
Battles:
70,949
$5.00 / $30.00−35%
DeepSeek V4 Pro1457.8
Evaluation details
Evaluated model:
DeepSeek-V4-Pro
Confidence interval:
1453.9–1461.7
Battles:
57,572
$1.32 / $3.96—
DeepSeek V4 Flash1435.7
Evaluation details
Evaluated model:
DeepSeek-V4-Flash
Confidence interval:
1431.6–1439.7
Battles:
51,944
$0.30 / $1.20—
Claude Opus 4.71493.7
Evaluation details
Evaluated model:
Claude Opus 4.7
Confidence interval:
1490.0–1497.5
Battles:
66,140
$5.00 / $25.00−35%
GPT-5.4 mini1447.2
Evaluation details
Evaluated model:
GPT-5.4 mini (High)
Confidence interval:
1443.4–1450.9
Battles:
63,432
$0.75 / $4.50−20%
GPT-5.41465.0
Evaluation details
Evaluated model:
GPT-5.4
Confidence interval:
1461.4–1468.7
Battles:
67,856
$2.50 / $15.00−20%
GPT-5.3 CodexNot matched$1.75 / $14.00−20%
Gemini 3.1 ProNot matched$2.00 / $12.00−45%
Claude Sonnet 4.61472.0
Evaluation details
Evaluated model:
Claude Sonnet 4.6
Confidence interval:
1468.4–1475.5
Battles:
70,916
$3.00 / $15.00−35%
GLM-51457.2
Evaluation details
Evaluated model:
GLM-5
Confidence interval:
1453.0–1461.5
Battles:
29,255
$1.00 / $3.20—
Kimi K2.51449.7
Evaluation details
Evaluated model:
Kimi K2.5 (Thinking)
Confidence interval:
1446.5–1453.0
Battles:
74,682
$0.60 / $3.00—
Gemini 3 Flash1472.7
Evaluation details
Evaluated model:
Gemini 3 Flash
Confidence interval:
1468.4–1477.1
Battles:
31,301
$0.50 / $3.00−45%
Claude Haiku 4.51414.5
Evaluation details
Evaluated model:
Claude Haiku 4.5 (20251001)
Confidence interval:
1412.1–1416.9
Battles:
154,063
$1.00 / $5.00−35%
GPT-5 CodexNot matched$1.25 / $10.00−20%
GPT-51434.7
Evaluation details
Evaluated model:
GPT-5 (High)
Confidence interval:
1430.2–1439.2
Battles:
31,718
$1.25 / $10.00−20%
Gemini 2.5 Flash-LiteNot matched$0.10 / $0.40−45%
Gemini 2.5 Flash1409.2
Evaluation details
Evaluated model:
Gemini 2.5 Flash
Confidence interval:
1406.8–1411.7
Battles:
125,457
$0.30 / $2.50−45%
OpenAI o31431.7
Evaluation details
Evaluated model:
o3 (2025-04-16)
Confidence interval:
1428.0–1435.3
Battles:
58,901
$2.00 / $8.00Not verified
GPT-4.11414.6
Evaluation details
Evaluated model:
GPT-4.1 (2025-04-14)
Confidence interval:
1410.9–1418.4
Battles:
50,298
$2.00 / $8.00−20%
GPT-4.1 mini1382.8
Evaluation details
Evaluated model:
GPT-4.1 mini (2025-04-14)
Confidence interval:
1378.5–1387.1
Battles:
38,696
$0.40 / $1.60−20%
Gemini 2.5 Pro1445.4
Evaluation details
Evaluated model:
Gemini 2.5 Pro
Confidence interval:
1442.9–1447.9
Battles:
125,307
$1.25 / $10.00−45%
GPT-4o mini1317.6
Evaluation details
Evaluated model:
GPT-4o mini (2024-07-18)
Confidence interval:
1314.0–1321.1
Battles:
68,705
$0.15 / $0.60−35%
GPT-4oNot matched$2.50 / $10.00−20%
Qwen3.8 Max1482.4
Evaluation details
Evaluated model:
Qwen3.8-Max
Confidence interval:
1477.6–1487.2
Battles:
31,915
$2.00 / $6.00—Not verified
MiMo V2.6 Pro1479.7
Evaluation details
Evaluated model:
MiMo-V2.6-Pro
Confidence interval:
1470.3–1489.1
Battles:
4,037
$0.435 / $0.87—Not verified

Text prices: USD per 1M tokens (input / output), standard base tier. Image examples show output cost; input may be billed separately. Site discounts come from the current billing configuration. Dates cite official announcements or API release notes, including previews; they are not listing dates on this site.

Leaderboard: Text · style controlled · Published:

Official Arena data; selected models and scores rounded to one decimal. Higher is better within the same leaderboard and publication date; scores are not percentages or comparable across tasks. Evaluation settings may differ from your API requests. Unmatched versions are left blank.

Source: Arena · License: CC BY 4.0

What is a chat & reasoning API?

An LLM API returns text (and increasingly tool calls and structured output) from a prompt. The models differ in reasoning depth, context window, speed and price; a gateway lets you route each request to the model that fits instead of integrating four vendors.

How to choose

Hard reasoning and coding

Claude Opus 5 / Opus 4.8, GPT-6 Astra, GPT-5.5, Gemini 3.1 Pro. Highest output price; reserve for tasks where quality changes the result.

Everyday assistants

Claude Sonnet 5 / 4.6, GPT-5.6 Sol, Gemini 3 Flash, DeepSeek V4 Flash, Kimi K3. Best price-to-quality for summaries, drafting, extraction.

High volume, low cost

Claude Haiku 4.5, GPT-5.4 mini, Gemini 2.5 Flash-Lite, GLM-5. Fractions of a cent per request; pair with caching.

Long context

Gemini 3.1 Pro and GPT-4.1 (1M tokens), Claude (200K). Check the context column before loading a whole repo.

One request shape for every model

Open a model's detail page to copy its API model ID.

cURL
curl https://api.crazyrouter.com/v1/chat/completions \
  -H "Authorization: Bearer $CRAZYROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemini-2.5-flash-lite","messages":[{"role":"user","content":"Summarize this PR in three bullets."}]}'

FAQ

What base URL do I use for chat models?

https://api.crazyrouter.com/v1 for OpenAI-compatible chat completions; Claude models also accept /v1/messages and Gemini models /v1beta.

Is the price the same as the vendor's?

No — the table shows the Crazyrouter price, which already includes the discount against each vendor's list price; the list price is shown for comparison.

Do streaming, tool calling and JSON mode work?

Yes, they behave as the upstream model documents them. Server-sent events are on by default when you pass stream: true.

Can I try a model before paying?

Yes — the newest chat models run once free with no account, from the Try-it box on their model page (no card). Every other model runs in the Playground after signup, which includes $2 of free credit; the API itself is pay as you go.

Which chat model is the cheapest on Crazyrouter?

Gemini 2.5 Flash-Lite at $0.055 per 1M input tokens as of the latest pricing sync; see the table for output prices.

Guides and comparisons

Podcast

AI Dev Tools: The Crazyrouter Podcast

Short episodes for developers on AI APIs, gateways, cost control and coding tools.

Latest episodes

  • EP196 · Oct 29, 2026 · 6:53

    AI API Security Posture — Turn Controls Into Continuous Evidence

    A practical guide to AI API security posture: inventory exposure, verify controls continuously, protect keys and data, constrain tools, and turn security assumptions into evidence that operators can act on.

  • EP195 · Oct 28, 2026 · 7:21

    AI API Policy Drift — Detect When Production Rules Stop Matching Intent

    A practical guide to policy drift in AI APIs: version routing and safety rules, detect stale enforcement, compare intended and observed behavior, and restore control before small gaps become incidents.

Episodes with written show notes →