surp/free — real responses, zero payment, treasury-sponsored.
surp/free gives users genuinely free LLM inference through the same OpenAI-compatible endpoint. There is no x402 payment challenge and no API key. The surp treasury pays the upstream model cost, with strict daily budgets, per-IP limits, cheap-model guardrails, and automatic fallback.
curl -X POST https://surp.ivc.lol/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{"model":"surp/free",
"messages":[{"role":"user","content":"Explain x402 in one sentence"}],
"max_tokens":64,
"stream":false}'
Limits: 10 requests/IP/day · 128 max output tokens · non-streaming · no tool calls · availability is best-effort.
| requests served today | 11 / 100 |
| tokens served today | 3,401 / 20,000 |
| failed upstream attempts | 20 |
| maximum sponsored model price | $0.0500 / 1M tokens |
| fallback attempts per request | up to 3 live models |
These are real, liquid Surplus marketplace models currently below the configured treasury ceiling. The router tries the cheapest first and falls back when a seller is unavailable.
| # | model | USD/1M | healthy sellers | requests 24h | volume 24h |
|---|---|---|---|---|---|
| 1 | jev-1.13 | $0.0004 | 5 | 280,081 | 291,246 |
| 2 | openai-gpt-oss-20b | $0.0011 | 437 | 570 | 154 |
| 3 | gemma-3-4b-it | $0.0015 | 142 | 284 | 1,524 |
| 4 | gemma-3-12b-it | $0.0020 | 83 | 391 | 1,928 |
| 5 | ministral-3-3b-instruct | $0.0020 | 180 | 507 | 1,961 |
| 6 | glm-4.7-flash | $0.0024 | 201 | 4,439 | 700,761 |
| 7 | deepseek-v4-flash | $0.0024 | 64 | 64,074 | 36,721,431 |
| 8 | gemma-3-27b-it | $0.0028 | 212 | 4,142 | 3,087 |
| 9 | gemma-4-26b-a4b-it | $0.0028 | 459 | 3,159 | 22,342 |
| 10 | deepseek-v4.1-flash | $0.0029 | 101 | 1,751,088 | 201,223,118 |
| 11 | gemini-2.5-flash-image | $0.0033 | 384 | 4 | 4 |
| 12 | qwen3-coder-30b-a3b-instruct | $0.0035 | 182 | 792 | 2,584 |
| 13 | qwen3-32b | $0.0036 | 185 | 669 | 927 |
| 14 | gemma-4-31b | $0.0036 | 168 | 720 | 2,304 |
| 15 | openai-gpt-oss-120b | $0.0037 | 239 | 4,310 | 42,075 |
| 16 | nvidia-nemotron-3-nano-30b-a3b | $0.0037 | 251 | 676 | 4,374 |
| 17 | qwen3-235b-a22b-2507 | $0.0040 | 204 | 6,508 | 12,348 |
| 18 | glm-5.2 | $0.0040 | 245 | 97,152 | 6,811,032 |
| 19 | ministral-3-14b-instruct | $0.0041 | 181 | 464 | 1,815 |
| 20 | ministral-3-8b-instruct | $0.0042 | 182 | 493 | 1,644 |
| 21 | gemma-4-31b-it | $0.0043 | 642 | 14,826 | 4,877,207 |
| 22 | glm-4.5-air | $0.0052 | 113 | 3,672 | 38,535 |
| 23 | nvidia-nemotron-nano-9b-v2 | $0.0057 | 177 | 64 | 1,535 |
| 24 | gemma-4-e2b | $0.0059 | 170 | 72 | 1,532 |
| 25 | qwen3.8-flash | $0.0061 | 139 | 15,390 | 850,280 |
| 26 | deepseek-v3.2 | $0.0067 | 148 | 240,374 | 11,392,116 |
| 27 | openai-gpt-oss-safeguard-20b | $0.0068 | 180 | 270 | 5,841 |
| 28 | kimi-k2.6 | $0.0070 | 172 | 3,105 | 1,075,250 |
| 29 | nvidia-nemotron-3-super-120b | $0.0080 | 182 | 1,853 | 13,013 |
| 30 | nvidia-nemotron-nano-12b-v2 | $0.0089 | 171 | 88 | 2,004 |
| 31 | qwen3-coder-next | $0.0091 | 140 | 1,980 | 18,574 |
| 32 | minimax-m2.5 | $0.0095 | 176 | 4,742 | 360,838 |
| 33 | glm-5.3-flash | $0.0097 | 163 | 238,298 | 35,487,445 |
| 34 | qwen3-next-80b-a3b-instruct | $0.0103 | 235 | 947 | 10,920 |
| 35 | minimax-m2 | $0.0118 | 155 | 705 | 21,697 |
| 36 | qwen3-coder | $0.0129 | 192 | 3,107 | 78,949 |
| 37 | mistral-small-3.2-24b-instruct | $0.0138 | 65 | 768 | 16,518 |
| 38 | gpt-5.6-luna | $0.0143 | 57 | 277,797 | 19,598,769 |
| 39 | openai-gpt-oss-safeguard-120b | $0.0144 | 196 | 171 | 22,464 |
| 40 | glm-4.7 | $0.0144 | 328 | 6,807 | 6,792,171 |
| 41 | minimax-m2.1 | $0.0148 | 184 | 1,244 | 9,522 |
| 42 | gpt-6-luna | $0.0150 | 45 | 544,012 | 48,454,901 |
| 43 | deepseek-v4-flash-0731 | $0.0158 | 123 | 53,475 | 7,502,673 |
| 44 | palmyra-vision-7b | $0.0168 | 174 | 37 | 199 |
| 45 | gemini-3.1-flash-lite | $0.0173 | 553 | 145,842 | 3,075,960 |
| 46 | deepseek-v3.1 | $0.0189 | 138 | 26,200 | 487,176 |
| 47 | qwen3-vl-235b-a22b-instruct | $0.0209 | 188 | 20,060 | 87,498 |
| 48 | glm-4.6 | $0.0216 | 336 | 15,799 | 4,218,468 |
| 49 | gpt-4o-mini | $0.0221 | 405 | 9,158 | 107,359 |
| 50 | minimax-m3 | $0.0224 | 32 | 4,834 | 4,164,444 |
| model | free requests | tokens |
|---|---|---|
| openai-gpt-oss-20b | 10 | 3,319 |
| glm-5.3-flash | 1 | 82 |
OmniRoute is an MIT-licensed AI gateway with an unusually careful free-tier catalog. We incorporated the parts that improve honesty and transparency:
Source: OmniRoute release/v3.8.50, curated 2026-07-22, MIT License. Snapshot stored with license attribution in this repository.
| providers represented | 81 |
| permanently free, uncapped providers | 13 (listed, never summed) |
| ToS generally ok | 49 |
| ToS ambiguous | 38 |
| ToS caution | 334 |
| ToS avoid | 93 |
| catalog entries after excluding avoid | 426 |
Because "free" often means free for one developer's account—not free to resell through a public proxy. OmniRoute itself surfaces these ToS risks. surp will only integrate a third-party free provider into public routing when its terms explicitly allow proxying/resale or we have written permission. Until then, the catalog is informational and our public free requests are treasury-sponsored.
related: live system status · cache-aware routing · reward proposal · token-gated access · API docs