Esc to close · ⌘K / Ctrl-K opens search anywhere
The cheapest model, the snappiest model and the fastest-streaming model are usuallythree different models — and the winner flips with your workload. Pick your use case; we rank the catalog by what that workload actually feels:₹ per task, time to first token andsustained throughput, measured through the production gateway (sweep of 2026-07-02), not read off datasheets.
Agent loops make dozens of short, tool-calling turns per task — time to first token dominates how fast the agent feels, throughput matters for big diffs, and cost adds up across the loop. Reasoning support is required.
| # | Model | Best route | ₹/Mtok (blended) | First token | tok/s |
|---|---|---|---|---|---|
| 1 | gpt-oss-120bopen weights🇮🇳 India route🔧 tools | fireworks | ₹39 | 1.2s | 469 |
| 2 | qwen3-32bopen weights🔧 tools | groqBYOK | ₹22 | 2.1s | 382 |
| 3 | gpt-oss-20bopen weights🇮🇳 India route🔧 tools | krutrim | ₹24 | 4.7s | 606 |
| 4 | glm-4.7open weights🔧 tools | openrouterBYOK | ₹136 | 1.9s | 68 |
| 5 | gemini-2.5-flash🔧 tools | openrouter | ₹187 | 2.2s | 132 |
| 6 | glm-4.7-flashopen weights🔧 toolsprice-ranked | zhipuBYOK | free | — | — |
| 7 | glm-4.5-flashopen weights🔧 toolsprice-ranked | zhipuBYOK | free | — | — |
| 8 | nemotron-3.5-lightning-30b-a3bopen weights🔧 toolsprice-ranked | nvidiaBYOK | free | — | — |
Blended ₹/Mtok = cheapest route at a 1:3 input:output token mix (generation-dominant). First token and tok/s are the best measured route per model. Rankings are a weighted percentile score per lens — details in the methodology below.
Click a metric column to sort by that lens. The same model often wins one and loses another.
| Model | Routes | ₹/Mtok ↕ | First token ↕ | tok/s ↕ |
|---|---|---|---|---|
| glm-4.7-flashreasoning🔧 tools | zhipuopenrouter | free | — | — |
| glm-4.5-flashreasoning🔧 tools | zhipu | free | — | — |
| nemotron-3.5-lightning-30b-a3breasoning🔧 tools | nvidia | free | — | — |
| nemotron-nano-omni-30breasoning | dgxspark2 🇮🇳 | ₹1.8 | — | — |
| qwen2.5-coder-7b | bharatrouter 🇮🇳 | ₹3.5 | — | — |
| qwen2.5-7b-instruct🔧 tools | bharatrouter 🇮🇳 | ₹3.5 | — | — |
| qwen3-8breasoning🔧 tools | bharatrouter 🇮🇳 | ₹3.5 | — | — |
| nemotron-super-120breasoning🔧 tools | dgxspark 🇮🇳 | ₹3.5 | — | — |
| qwen2.5-vl-7b-instruct | bharatrouter 🇮🇳 | ₹5.3 | — | — |
| gemma-4-e4b-it🔧 tools | krutrim 🇮🇳 | ₹6.8 | — | — |
| qwen3.5-9breasoning🔧 tools | krutrim 🇮🇳 | ₹6.8 | — | — |
| llama-3.1-8b-instruct🔧 tools | openroutergroqfireworks | ₹7.0 | 428ms groq | 548 |
| glm-4-32b-0414-128k🔧 tools | zhipu | ₹10 | — | — |
| command-r7b🔧 tools | cohere | ₹12 | — | — |
| qwen3-32breasoning🔧 tools | groqopenrouterfireworks | ₹22 | 2.1s groq | 382 |
| gemma-4-26b-a4b-it🔧 tools | krutrim 🇮🇳 | ₹23 | — | — |
| qwen3.6-35b-a3breasoning🔧 tools | krutrim 🇮🇳 | ₹23 | — | — |
| gpt-oss-20breasoning🔧 tools | krutrim 🇮🇳groqfireworks | ₹24 | 4.7s krutrim | 606 |
| deepseek-v4-flashreasoning🔧 tools | deepseek | ₹24 | — | — |
| devstral-small🔧 tools | mistral | ₹24 | — | — |
| llama-3.3-70b🔧 tools | groqopenrouterfireworks | ₹26 | 2.0s openrouter | 218 |
| gemma-4-31b-it🔧 tools | krutrim 🇮🇳 | ₹27 | — | — |
| llama-4-scout🔧 tools | groqfireworks | ₹28 | — | — |
| glm-4.6v-flashxreasoning🔧 tools | zhipu | ₹30 | — | — |
| glm-4.7-flashxreasoning🔧 tools | zhipu | ₹30 | — | — |
| gemini-2.5-flash-litereasoning🔧 tools | gemini | ₹31 | — | — |
| gpt-oss-120breasoning🔧 tools | krutrim 🇮🇳groqbasetenfireworks | ₹39 | 1.2s fireworks | 469 |
| glm-5.3-flashreasoning🔧 tools | krutrim 🇮🇳openrouter | ₹40 | — | — |
| mistral-small🔧 tools | mistral | ₹47 | — | — |
| command-r🔧 tools | cohere | ₹47 | — | — |
| gpt-4o-mini🔧 tools | openai | ₹47 | — | — |
| gpt-4o-mini-search-preview🔧 tools | openai | ₹47 | — | — |
| nemotron-superreasoning🔧 tools | baseten | ₹61 | — | — |
| glm-4.5-airreasoning🔧 tools | zhipuopenrouterfireworks | ₹65 | — | — |
| glm-4.6vreasoning🔧 tools | zhipuopenrouter | ₹72 | — | — |
| codestral🔧 tools | mistral | ₹72 | — | — |
| deepseek-v4-proreasoning🔧 tools | deepseekbasetenfireworks | ₹74 | — | — |
| deepseek-v3🔧 tools | openrouter | ₹81 | 16.3s openrouter | 38 |
| gpt-5.6-lunareasoning🔧 tools | openai | ₹91 | — | — |
| sonar🔧 tools | perplexity | ₹96 | — | — |
| grok-code-fast-1reasoning🔧 tools | xai | ₹113 | — | — |
| gemini-3.1-flash-litereasoning🔧 tools | gemini | ₹114 | — | — |
| mistral-largereasoning🔧 tools | mistral | ₹120 | — | — |
| glm-4.7reasoning🔧 tools | basetenzhipuopenrouter | ₹136 | 1.9s openrouter | 68 |
| gpt-5-minireasoning🔧 tools | openai | ₹150 | 10.2s openai | 63 |
| glm-5reasoning🔧 tools | basetenzhipuopenrouter | ₹153 | 12.2s zhipu | 53 |
| glm-4.6reasoning🔧 tools | zhipuopenrouter | ₹156 | — | — |
| grok-build-0.1reasoning🔧 tools | xaiopenrouter | ₹168 | — | — |
| glm-4.5reasoning🔧 tools | zhipuopenrouter | ₹173 | — | — |
| kimi-k2.5reasoning🔧 tools | basetenopenrouter | ₹173 | — | — |
| kimi-k2🔧 tools | groqopenrouter | ₹180 | 688ms groq | 178 |
| nemotron-ultrareasoning🔧 tools | baseten | ₹187 | — | — |
| gemini-2.5-flashreasoning🔧 tools | geminiopenrouter | ₹187 | 2.2s openrouter | 132 |
| gemini-3.5-flash-litereasoning🔧 tools | gemini | ₹187 | — | — |
| grok-4.3reasoning🔧 tools | xaiopenrouter | ₹210 | — | — |
| grok-4.20reasoning🔧 tools | xaiopenrouter | ₹210 | — | — |
| glm-5.2reasoning🔧 tools | basetenzhipuopenrouterfireworks | ₹238 | 5.7s zhipu | 68 |
| glm-5.1reasoning🔧 tools | basetenzhipuopenrouter | ₹242 | — | — |
| kimi-k2.7-codereasoning🔧 tools | fireworksmoonshotbasetenopenrouter | ₹270 | 19.6s openrouter | 39 |
| gemini-3.7-flashreasoning🔧 tools | gemini | ₹288 | — | — |
| kimi-k2.6reasoning🔧 tools | fireworksmoonshotbasetenopenrouter | ₹311 | — | — |
| glm-5-turboreasoning🔧 tools | zhipu | ₹317 | — | — |
| glm-5v-turboreasoning🔧 tools | zhipuopenrouter | ₹317 | — | — |
| glm-5.3reasoning🔧 tools | fireworkszhipu | ₹350 | — | — |
| glm-4.5-airxreasoning🔧 tools | zhipu | ₹351 | — | — |
| claude-haiku-4.5reasoning🔧 tools | anthropicopenrouter | ₹384 | 16.8s openrouter | 102 |
| qwen3.8-maxreasoning🔧 tools | alibabaopenrouter | ₹480 | — | — |
| pixtral-large🔧 tools | mistral | ₹480 | — | — |
| grok-4.6reasoning🔧 tools | xai | ₹480 | — | — |
| grok-4.5reasoning🔧 tools | xai | ₹480 | — | — |
| gemini-3.6-flashreasoning🔧 tools | gemini | ₹576 | — | — |
| mistral-medium-3.5🔧 tools | mistral | ₹576 | — | — |
| kimi-k2.7-code-highspeedreasoning🔧 tools | moonshot | ₹622 | — | — |
| sonar-reasoning-proreasoning🔧 tools | perplexity | ₹624 | — | — |
| sonar-deep-researchreasoning🔧 tools | perplexity | ₹624 | — | — |
| gemini-3.5-flashreasoning🔧 tools | gemini | ₹684 | — | — |
| glm-4.5-xreasoning🔧 tools | zhipu | ₹693 | — | — |
| gemini-2.5-proreasoning🔧 tools | geminiopenrouter | ₹750 | — | — |
| gpt-5reasoning🔧 tools | openai | ₹750 | — | — |
| claude-sonnet-5reasoning🔧 tools | anthropicopenrouter | ₹768 | — | — |
| command-a🔧 tools | cohere | ₹780 | — | — |
| gpt-4o-search-preview🔧 tools | openai | ₹780 | — | — |
| gpt-5.6-terrareasoning🔧 tools | openai | ₹912 | — | — |
| kimi-k3reasoning🔧 tools | fireworksmoonshotopenrouterbaseten | ₹1152 | — | — |
| sonar-pro🔧 tools | perplexity | ₹1152 | — | — |
| claude-opus-4.8reasoning🔧 tools | anthropicopenrouter | ₹1920 | — | — |
| claude-opus-5reasoning🔧 tools | anthropic | ₹1920 | — | — |
| gpt-5.6-solreasoning🔧 tools | openai | ₹2280 | — | — |
| gpt-5.6reasoning🔧 tools | openai | ₹2280 | — | — |
| claude-fable-5reasoning🔧 tools | anthropicopenrouter | ₹3840 | — | — |
| gpt-6-astrareasoning🔧 tools | openai | ₹3840 | — | — |
Perf numbers are medians from a multi-run streamed sweep through the production gateway on 2026-07-02: every (model × provider) route gets the same ~300-token prompt, rounds interleaved across hosts so no provider owns a time-of-day advantage. First token counts reasoning tokens (it's what you see). tok/s is the post-first-token decode rate. Models not yet swept show “—” and rank on price with a neutral perf score. Routes we could not measure are listed openly in theAPI response(9 skipped this sweep), never silently dropped. Numbers refresh with each sweep; live per-route health is on /models.
Agents get this same chooser as JSON:GET /v1/compare/models — rankings, per-route pricing, measured perf and live failure rates, no auth required.