LLM gateways and observability for small business buyers
Asked as “LLM gateway”, and as “LLM observability and routing platform”, on behalf of a small B2B company. 61 first choices recorded across the direct, paraphrase, budget and scale prompts, fourteen models each.
Standing · first-choice share
36%
Contested · LiteLLM 33%
36Portkey33LiteLLM10Helicone21others
36% of first choices, contested.
Since October 2026↗new leaderNew leader since October 2026: Portkey (36%) replaces LiteLLM (31% then, 33% now), 3 points clear, inside the 11-point floor.Portkey leads at 36%, replacing LiteLLM, which led at 31% and stands at 33% now: 3 points clear, inside the floor, so the swap reads as unsettled.
By buyer segment
The same question asked on behalf of a different buyer. Each standing is computed within its segment; they sit side by side and are never added together.
Small business
Portkey36%Contested↗new leaderNew leader since October 2026: Portkey (36%) replaces LiteLLM (31% then, 33% now), 3 points clear, inside the 11-point floor.This pageMid-market
LiteLLM36%Contested↗new leaderNew leader since October 2026: LiteLLM (36%) replaces Portkey (33% then, 32% now), 4 points clear, inside the 11-point floor.Standing →Enterprise
Bifrost23%Contested↗new leaderNew leader since October 2026: Bifrost (23%) replaces Portkey (26% then, 14% now), 9 points clear, inside the 11-point floor.Standing →
Portkey36%Contested↗new leaderNew leader since October 2026: Portkey (36%) replaces LiteLLM (31% then, 33% now), 3 points clear, inside the 11-point floor.This pageMid-marketLiteLLM36%Contested↗new leaderNew leader since October 2026: LiteLLM (36%) replaces Portkey (33% then, 32% now), 4 points clear, inside the 11-point floor.Standing →EnterpriseBifrost23%Contested↗new leaderNew leader since October 2026: Bifrost (23%) replaces Portkey (26% then, 14% now), 9 points clear, inside the 11-point floor.Standing →The standing
Share is the count of first choices across the direct, paraphrase, budget and scale prompts, over all fourteen models, for a small B2B company. Ordered by share.
ProductFirst-choice shareNegative rateLabelsQuadrantSince October 2026
01Portkey36%13%55endorsed leader▲+8Since October 2026: 28% → 36%, +8 points. Inside the 11-point floor: within noise.28% → 36%
Portkey36%13%55endorsed leader▲+8Since October 2026: 28% → 36%, +8 points. Inside the 11-point floor: within noise.28% → 36%02LiteLLM33%21%71endorsed leader▲+2Since October 2026: 31% → 33%, +2 points. Inside the 11-point floor: within noise.31% → 33%
LiteLLM33%21%71endorsed leader▲+2Since October 2026: 31% → 33%, +2 points. Inside the 11-point floor: within noise.31% → 33%03
Helicone10%7%28accepted challenger▲+5Since October 2026: 5% → 10%, +5 points. Inside the 11-point floor: within noise.5% → 10%
Helicone10%7%28accepted challenger▲+5Since October 2026: 5% → 10%, +5 points. Inside the 11-point floor: within noise.5% → 10%04
OpenRouter7%19%54accepted challenger▼−14Since October 2026: 21% → 7%, −14 points. Past the 11-point floor: movement.21% → 7%
OpenRouter7%19%54accepted challenger▼−14Since October 2026: 21% → 7%, −14 points. Past the 11-point floor: movement.21% → 7%05
Cloudflare AI Gateway3%9%32accepted challenger▼−4Since October 2026: 7% → 3%, −4 points. Inside the 11-point floor: within noise.7% → 3%
Cloudflare AI Gateway3%9%32accepted challenger▼−4Since October 2026: 7% → 3%, −4 points. Inside the 11-point floor: within noise.7% → 3%06
Langfuse3%0%13accepted challenger▲+3Since October 2026: 0% → 3%, +3 points. Inside the 11-point floor: within noise.0% → 3%
Langfuse3%0%13accepted challenger▲+3Since October 2026: 0% → 3%, +3 points. Inside the 11-point floor: within noise.0% → 3%07
Vercel AI Gateway2%19%16accepted challenger=heldSince October 2026: 2% → 2%, ±0 points. Inside the 11-point floor: within noise.2% → 2%
Vercel AI Gateway2%19%16accepted challenger=heldSince October 2026: 2% → 2%, ±0 points. Inside the 11-point floor: within noise.2% → 2%Show the two products at 0%, ordered by negative rate
09
Kong AI Gateway0%35%17criticized challenger=heldSince October 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise.0% → 0%
Kong AI Gateway0%35%17criticized challenger=heldSince October 2026: 0% → 0%, ±0 points. Inside the 11-point floor: within noise.0% → 0%08Bifrost0%9%22accepted challenger▼−2Since October 2026: 2% → 0%, −2 points. Inside the 11-point floor: within noise.2% → 0%
Bifrost0%9%22accepted challenger▼−2Since October 2026: 2% → 0%, −2 points. Inside the 11-point floor: within noise.2% → 0%The floor is 11 points of share, measured: how far the models move a leader on their own when the same questions are asked twice with nothing changed. A larger change is movement; a smaller one can't be told apart from repeat variation, and both are shown. Each index measures its own floor from its own repeat, over its own categories, so the floor differs from one index to the next. Both editions asked the same fourteen models. How the floor is measured
Bars are the share of first choices, 0 to 100Every product with at least 10 labels here. Every product name links to its product page.
For buyers
Head to headHead-to-head pages are published for mid-market buyers, so these open on mid-market figures, not small business.
All twenty-one head-to-head pages: the top seven products, each against each
LiteLLM vs Cloudflare AI GatewayLiteLLM vs OpenRouterLiteLLM vs LangfusePortkey vs BifrostPortkey vs HeliconePortkey vs Cloudflare AI GatewayPortkey vs OpenRouterPortkey vs LangfuseBifrost vs HeliconeBifrost vs Cloudflare AI GatewayBifrost vs OpenRouterBifrost vs LangfuseHelicone vs Cloudflare AI GatewayHelicone vs OpenRouterHelicone vs LangfuseCloudflare AI Gateway vs OpenRouterCloudflare AI Gateway vs LangfuseOpenRouter vs Langfuse
Recommended versus criticized
Every product with at least 10 labels here, on both axes. The 30% line names a quadrant, not the verdict above: that one needs more than 40%.
Criticized challengerCriticized default
Negative label rate →
010203040506070809Accepted challengerEndorsed leader
0%First-choice share → · lines at 30% share and 25% negative50%
Key
01Portkey36%
Portkey36%02LiteLLM33%
LiteLLM33%03Helicone10%
Helicone10%04OpenRouter7%
OpenRouter7%05Cloudflare AI Gateway3%
Cloudflare AI Gateway3%06Langfuse3%
Langfuse3%07Vercel AI Gateway2%
Vercel AI Gateway2%08Bifrost0%
Bifrost0%09Kong AI Gateway0%
Kong AI Gateway0%What they warned about
Three of fourteen models held their first choice under the paraphrase. Claude Haiku 4.5, GPT-5.4 mini, Grok 4.20, Mistral Small, DeepSeek V4 Flash, Llama 4 Maverick, Qwen 3.7 Flash, Kimi K2, GLM 4.7 FlashX, GPT-6 Luna and Muse Glimmer 30B changed. A high negative share on a product with few labels is a warning. A low share on a product with many labels is salience, not sentiment.
LiteLLM21%
15 of 71 labels negative · 12 of 14 models · 5 hard negative
“Be very cautious or avoid if not security-savvy... hundreds of publicly exposed LiteLLM instances worldwide have been exploited” Grok 4.20, negative prompt
OpenRouter19%
10 of 54 labels negative · 10 of 14 models · 3 hard negative
“for production workloads involving customer data, proprietary business logic, or regulated information, you should avoid it” DeepSeek V4 Flash, negative prompt
Portkey13%
7 of 55 labels negative · 6 of 14 models · 1 hard negative
“Avoid $49/mo managed gateways like Portkey/Helicone until your monthly LLM spend justifies the governance/observability premium.” Muse Glimmer 30B, budget prompt
Kong AI Gateway35%
6 of 17 labels negative · 4 of 14 models · 2 hard negative
“Avoid fully managed enterprise options (e.g., full Kong or TrueFoundry) unless you have existing infrastructure” Grok 4.20, direct prompt
What they cite
Fourteen of the fourteen models returned a source list in this edition: five return the pages they cited and nine the search results they consulted.
Sites the answers cite
74 of 84 answers in this category came back with a source list, from 14 of 14 models: citations where the model returns them, or the search results it consulted. 821 links across 188 sites, every framing counted. Ranked by the number of answers carrying the site or page. 38 of the 252 answers across every segment cited this index's own page for the category; the method page measures whether that reading tilts an answer.
41 answers · 63 citations · 12 models
35 answers · 76 citations · 12 models
vendor site · GitHub22 answers · 36 citations · 9 models
vendor site · OpenRouter20 answers · 36 citations · 8 models
publisher (Gane)20 answers · 21 citations · 9 models
vendor site · Requesty19 answers · 23 citations · 8 models
vendor site · Respan19 answers · 20 citations · 7 models
vendor site · Portkey17 answers · 22 citations · 7 models
vendor site · LLMGateway.io16 answers · 25 citations · 10 models
vendor site · CloudZero16 answers · 16 citations · 10 models
vendor site · LiteLLM15 answers · 25 citations · 7 models
vendor site · Dataiku13 answers · 13 citations · 8 models
Pages the answers cite
The ten pages named in the most answers, by full address. A page here is one the models returned with a recommendation, not one the index endorses.
22 answers · 9 models
20 answers · 9 models
16 answers · 10 models
15 answers · 7 models
14 answers · 7 models
13 answers · 8 models
11 answers · 6 models
9 answers · 5 models
9 answers · 6 models
9 answers · 5 models
Search against answers
Each company's standing in the answers beside its site's footprint in Google search, one row a site: the products the models named on it with their shares, and the share they add up to; monthly searches on Google, and DataForSEO's estimate of AI search demand (modeled from search signals, directional, not a count of queries to any assistant), for the most-searched of the company's and its products' names (the name is in each row's hover text); estimated monthly organic visits to the site; and its best position in Google's top ten for “best llm gateway”, “llm gateway”, “llm gateways”. US estimates from DataForSEO and Google's Ads Transparency Center. A small company's site, or a mid-sized company's site for its flagship, is marked company; a product on a large parent's site (Google, Microsoft) has no site figures. A column with no figures for this category is left out, and an empty cell means none were seen, not none exist. Two measurements side by side: neither is read as the cause of the other.
| Company and products | Share | Own site cited | Name searches, Google | AI search demand, est. | Organic visits | Google ads, last 30 days | Best Google position |
|---|---|---|---|---|---|---|---|
| Portkey portkey.ai | 36% | 156 | 5,400 | 131 | 5,999 | 0 | 5 |
| LiteLLM litellm.ai | 33% | 196 | 22,200 | 264 | 15,281 | 0 | 2 |
| Helicone helicone.ai | 10% | 50 | 1,600 | 899 | 2,439 | 0 | |
| OpenRouter openrouter.ai | 7% | 118 | 135,000 | 655 | 225,878 | 68 | 7 |
| Cloudflare parent site Cloudflare AI Gateway 3% | 3% | 96 | 550,000 | 10,897 | |||
| Langfuse langfuse.com | 3% | 48 | 18,100 | 216 | 8,530 | 4 | |
| Vercel AI Gateway vercel.com | 2% | 33 | 2,400 | 26 | 141,210 | 26 | |
| Anthropic anthropic.com | 2% | 15 | 2,657,653 | 3 | |||
| LangSmith langchain.com | 2% | 19 | 14,800 | 238 | 93,424 | 94 | |
| LLM Gateway llmgateway.io LLM Gateway 2%, LLMGateway.io 0% | 2% | 13 | 1,600 | 69 | 1,222 | 1 |
Google's top ten for the three searches hold 18 sites; 8 of them are among the sites the models cited here (dataiku.com, dev.to, getmaxim.ai, github.com, litellm.ai, llmgateway.io, openrouter.ai, portkey.ai). The first result for “best llm gateway” is reddit.com.
How they answered
Six framings of the same buying question, each sent to every model in a fresh session with search on. One row per model, so a row shows whether it held its answer under rewording, what it named when cost was the constraint, and what it argued against. Computed from the raw judge labels.ShowHide
How they answered
Six framings of the same buying question, each sent to every model in a fresh session with search on. One row per model, so a row shows whether it held its answer under rewording, what it named when cost was the constraint, and what it argued against. Computed from the raw judge labels.
ShowHideModelDirect“What is the best LLM gateway for a small B2B company?”Paraphrase“Which LLM observability and routing platform would you recommend to a small business that sells to other businesses?”Comparative“What are the top LLM gateways for a small team and how do they differ?”Budget-constrained“What is the best LLM gateway for a small company with a tight budget?”Scale-constrained“We are a 30 person company evaluating a LLM gateway. What should we look at?”Negative“Which LLM gateways should a small business avoid or be cautious about?”
Claude Haiku 4.5Anthropic, OpenAI APIPortkeyChangedagainst: TrueFoundry AI GatewayLiteLLM, OpenRouterHelicone, LiteLLMno first choiceagainst: LiteLLM, OpenRouter
One alternative
OpenRouterTwo alternatives
Helicone, LangfuseTwo alternatives
Cloudflare AI Gateway, HeliconeFour alternatives
Bifrost, Braintrust Gateway, Cloudflare AI Gateway, OpenRouterGPT-5.4 miniLiteLLM, OpenRouterLangfuseChangedLiteLLMLiteLLMagainst: Kong AI Gatewayno first choicenothing named
Two alternatives
Kong AI Gateway, PortkeyTwo alternatives
LiteLLM, OpenRouterFour alternatives
Cloudflare AI Gateway, Helicone, OpenRouter, PortkeyOne alternative
PortkeyGemini 3.5 FlashPortkeyPortkeyHeldLiteLLMCloudflare AI Gateway, LiteLLMLiteLLMnothing named
Four alternatives
Bifrost, Cloudflare AI Gateway, LiteLLM, OpenRouterThree alternatives
Helicone, Langfuse, LiteLLMFour alternatives
Bifrost, Cloudflare AI Gateway, OpenRouter, PortkeyTwo alternatives
Bifrost, OpenRouterSix alternatives
AWS Bedrock Gateway, Azure API Management, Cloudflare AI Gateway, Helicone, Langfuse, PortkeyPerplexity SonarPortkeyPortkeyHeldOpenRouteragainst: Bifrost, Kong AI GatewayLiteLLMagainst: Vercel AI Gatewayno first choiceagainst: Cloudflare AI Gateway, LiteLLM, OpenRouter, Portkey, Vercel AI Gateway
Four alternatives
Cloudflare AI Gateway, LiteLLM, OpenRouter, Vercel AI GatewayFour alternatives
Braintrust Gateway, Helicone, Langfuse, LiteLLMFour alternatives
Cloudflare AI Gateway, LiteLLM, Portkey, Vercel AI GatewayTwo alternatives
Cloudflare AI Gateway, OpenRouterGrok 4.20LiteLLM, Portkeyagainst: Bifrost, Cloudflare AI Gateway, Kong AI Gateway, TrueFoundry AI GatewayPortkeyChangedagainst: Datadog LLM Observability, LangSmithLiteLLMagainst: Apache APISIX, Braintrust Gateway, Inworld, Kong AI GatewayLiteLLMagainst: Portkeyno first choiceagainst: AWS, Apigee, Braintrust Gateway, Cloudflare AI Gateway, Helicone, Kong AI Gateway, LiteLLM, OpenRouter, Orq.ai, Portkey, Zuplo
Two alternatives
OpenRouter, RespanFour alternatives
FastRouter, Helicone, Langfuse, RespanFour alternatives
Bifrost, Helicone, OpenRouter, PortkeyFour alternatives
Bifrost, Helicone, Langfuse, OpenRouterFour alternatives
Helicone, LangSmith, LiteLLM, PortkeyMistral SmallLiteLLMLiteLLM, PortkeyChangedLiteLLMLLM Gateway, LiteLLMHelicone, LangSmithagainst: LiteLLM, Ollama, OpenRouter
Three alternatives
Bifrost, OpenRouter, PortkeyTwo alternatives
Confident AI, HeliconeFive alternatives
Bifrost, Cloudflare AI Gateway, OpenRouter, Portkey, Vercel AI GatewayThree alternatives
Cloudflare AI Gateway, Helicone, OpenRouterThree alternatives
Arize, LiteLLM, OpenLLMDeepSeek V4 FlashLiteLLMPortkeyChangedLiteLLM, OpenRouterLiteLLMCloudflare AI Gateway, Portkeyagainst: LiteLLM, OpenRouter, Portkey
Three alternatives
Cloudflare AI Gateway, OpenRouter, PortkeyTwo alternatives
Langfuse, LiteLLMFour alternatives
Cloudflare AI Gateway, Helicone, Portkey AI Gateway, Vercel AI GatewayThree alternatives
Cloudflare AI Gateway, OpenRouter, PortkeyOne alternative
LiteLLMLlama 4 MaverickPortkeyno first choiceChangedLiteLLMPortkeyno first choiceagainst: LiteLLM
Four alternatives
Bifrost, LiteLLM, OpenRouter, RequestyFive alternatives
Bifrost, Cloudflare AI Gateway, Kong AI Gateway, Merge Gateway, Vercel AI GatewayThree alternatives
LiteLLM, OpenRouter, RequestyQwen 3.7 FlashPortkeyHelicone, PortkeyChangedLiteLLMLiteLLMHelicone, Portkeyagainst: LiteLLMagainst: LiteLLM, OpenRouter, Portkey
Two alternatives
LiteLLM, OpenRouterOne alternative
LiteLLMThree alternatives
Helicone, Portkey, Vercel AI SDKTwo alternatives
Cloudflare AI Gateway, OpenRouterTwo alternatives
LangSmith, Vercel AI SDKKimi K2PortkeyHeliconeChangedLiteLLM, Portkeyagainst: Kong AI GatewayLiteLLMOpenRouter, Portkeyagainst: LLM Gateway, LiteLLM, OpenRouter
Two alternatives
LiteLLM, OpenRouterTwo alternatives
Langfuse, OpenRouterTwo alternatives
Bifrost, OpenRouterThree alternatives
Free LLM Gateway, LLM Gateway, OpenRouterOne alternative
LiteLLMGLM 4.7 FlashXOpenRouterPortkeyChangedOpenRouterLiteLLMno first choiceagainst: Helicone AI Gateway, LiteLLM, Martian Gateway, OpenRouter, Vercel AI Gateway
Three alternatives
LiteLLM, Portkey, Vercel AI GatewayThree alternatives
Helicone, Langfuse, OpenRouterSix alternatives
Amazon Bedrock, Bifrost, Braintrust Gateway, Kong AI Gateway, LiteLLM, PortkeyOne alternative
APIClawMiniMax M2.5PortkeyPortkeyHeldPortkeyLiteLLM, OpenRouterno first choiceagainst: LiteLLM
Two alternatives
LiteLLM, OpenRouterFour alternatives
Bifrost, Cloudflare AI Gateway, LiteLLM, OpenRouterTwo alternatives
LLMGateway.io, PortkeyGPT-6 LunaPortkeyHeliconeChangedCloudflare AI Gateway, Vercel AI Gatewayagainst: LiteLLMVercel AI Gatewayagainst: Portkeyno first choiceagainst: LiteLLM, OpenRouter
Three alternatives
Cloudflare AI Gateway, LiteLLM, OpenRouterThree alternatives
Langfuse, LiteLLM, PortkeyTwo alternatives
OpenRouter, PortkeyTwo alternatives
LiteLLM, OpenRouterMuse Glimmer 30BLiteLLM, PortkeyLangfuse, LiteLLMChangedLiteLLM, OpenRouterLiteLLMagainst: Helicone, Portkeyno first choiceagainst: LiteLLMagainst: LiteLLM, OpenRouter
Two alternatives
Cloudflare AI Gateway, OpenRouterThree alternatives
Arize AX, Confident AI, HeliconeThree alternatives
Bifrost, Cloudflare AI Gateway, Vercel AI GatewayTwo alternatives
LLMGateway.io, OpenRouterBold is the first choiceAlternatives are counted; the count opens them.What the answer argued against
The record
One row per call: the version string exactly as returned, whether the model searched, sources cited, and latency.
Eighty-four rows: every prompt, every model, every answer.
| Prompt | Model | Version string | Time (UTC) | Searched | Sources | Latency |
|---|---|---|---|---|---|---|
| Direct recommendation | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 19:25 | no | 0 | 5 s |
| Direct recommendation | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 13:23 | yes | 2 | 6 s |
| Direct recommendation | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 10:45 | yes | 10 | 18 s |
| Direct recommendation | Perplexity Sonar | sonar | 2026-11-01 14:58 | yes | 10 | 4 s |
| Direct recommendation | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 19:44 | yes | 5 | 9 s |
| Direct recommendation | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 19:11 | yes | 5 | 4 s |
| Direct recommendation | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 12:17 | yes | 24 | 27 s |
| Direct recommendation | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 16:23 | yes | 5 | 3 s |
| Direct recommendation | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 13:07 | yes | 5 | 33 s |
| Direct recommendation | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 09:48 | yes | 10 | 19 s |
| Direct recommendation | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 19:33 | yes | 23 | 36 s |
| Direct recommendation | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 17:38 | yes | 9 | 28 s |
| Direct recommendation | GPT-6 Luna | gpt-6-luna | 2026-11-01 11:21 | yes | 4 | 16 s |
| Direct recommendation | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 13:32 | yes | 14 | 23 s |
| Paraphrase | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 08:40 | yes | 10 | 8 s |
| Paraphrase | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 16:17 | yes | 3 | 5 s |
| Paraphrase | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 09:24 | no | 0 | 17 s |
| Paraphrase | Perplexity Sonar | sonar | 2026-11-01 19:38 | yes | 10 | 3 s |
| Paraphrase | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 14:54 | yes | 10 | 12 s |
| Paraphrase | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 08:58 | yes | 5 | 5 s |
| Paraphrase | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 16:22 | yes | 25 | 36 s |
| Paraphrase | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 12:39 | yes | 5 | 3 s |
| Paraphrase | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 19:22 | yes | 5 | 46 s |
| Paraphrase | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 14:12 | yes | 20 | 24 s |
| Paraphrase | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 16:05 | yes | 25 | 54 s |
| Paraphrase | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 12:19 | yes | 5 | 54 s |
| Paraphrase | GPT-6 Luna | gpt-6-luna | 2026-11-01 10:09 | yes | 2 | 13 s |
| Paraphrase | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 14:31 | yes | 15 | 21 s |
| Comparative | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 17:28 | yes | 18 | 11 s |
| Comparative | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 18:57 | yes | 7 | 8 s |
| Comparative | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 12:12 | yes | 11 | 24 s |
| Comparative | Perplexity Sonar | sonar | 2026-11-01 17:06 | yes | 10 | 4 s |
| Comparative | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 14:42 | yes | 5 | 12 s |
| Comparative | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 10:23 | yes | 5 | 8 s |
| Comparative | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 14:34 | yes | 24 | 50 s |
| Comparative | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 09:36 | yes | 5 | 3 s |
| Comparative | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 15:36 | no | 0 | 33 s |
| Comparative | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 09:39 | yes | 20 | 50 s |
| Comparative | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 11:22 | yes | 15 | 37 s |
| Comparative | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 19:39 | yes | 9 | 20 s |
| Comparative | GPT-6 Luna | gpt-6-luna | 2026-11-01 09:51 | yes | 8 | 35 s |
| Comparative | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 14:37 | yes | 15 | 26 s |
| Budget constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 09:16 | yes | 17 | 11 s |
| Budget constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 11:54 | yes | 3 | 4 s |
| Budget constrained | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 09:20 | yes | 4 | 23 s |
| Budget constrained | Perplexity Sonar | sonar | 2026-11-01 13:30 | yes | 10 | 3 s |
| Budget constrained | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 15:48 | yes | 5 | 8 s |
| Budget constrained | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 18:07 | yes | 5 | 5 s |
| Budget constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 19:35 | yes | 22 | 29 s |
| Budget constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 16:24 | yes | 5 | 3 s |
| Budget constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 18:12 | yes | 5 | 23 s |
| Budget constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 12:28 | yes | 10 | 15 s |
| Budget constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 16:38 | yes | 24 | 30 s |
| Budget constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 08:41 | yes | 13 | 50 s |
| Budget constrained | GPT-6 Luna | gpt-6-luna | 2026-11-01 13:13 | yes | 4 | 22 s |
| Budget constrained | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 14:55 | yes | 15 | 21 s |
| Scale constrained | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 08:45 | no | 0 | 7 s |
| Scale constrained | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 16:53 | no | 0 | 5 s |
| Scale constrained | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 16:35 | no | 0 | 16 s |
| Scale constrained | Perplexity Sonar | sonar | 2026-11-01 11:55 | yes | 10 | 6 s |
| Scale constrained | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 17:04 | yes | 5 | 15 s |
| Scale constrained | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 17:47 | no | 0 | 9 s |
| Scale constrained | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 17:24 | yes | 15 | 32 s |
| Scale constrained | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 13:01 | yes | 5 | 3 s |
| Scale constrained | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 10:34 | no | 0 | 30 s |
| Scale constrained | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 17:25 | yes | 15 | 28 s |
| Scale constrained | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 13:51 | yes | 10 | 46 s |
| Scale constrained | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 14:38 | no | 0 | 15 s |
| Scale constrained | GPT-6 Luna | gpt-6-luna | 2026-11-01 14:35 | yes | 1 | 18 s |
| Scale constrained | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 18:46 | yes | 15 | 18 s |
| Negative framing | Claude Haiku 4.5 | claude-haiku-4-5-20251001 | 2026-11-01 19:38 | yes | 17 | 10 s |
| Negative framing | GPT-5.4 mini | gpt-5.4-mini-2026-03-17 | 2026-11-01 16:44 | yes | 5 | 6 s |
| Negative framing | Gemini 3.5 Flash | gemini-3.6-flash (asked for gemini-3.5-flash; the API answered as this version) | 2026-11-01 13:59 | no | 0 | 22 s |
| Negative framing | Perplexity Sonar | sonar | 2026-11-01 14:16 | yes | 10 | 5 s |
| Negative framing | Grok 4.20 | spacexai/grok-4.20-non-reasoning via xai | 2026-11-01 15:46 | yes | 20 | 20 s |
| Negative framing | Mistral Small | mistral/mistral-small via mistral | 2026-11-01 12:22 | yes | 5 | 4 s |
| Negative framing | DeepSeek V4 Flash | deepseek/deepseek-v4-flash via deepinfra | 2026-11-01 09:20 | yes | 24 | 31 s |
| Negative framing | Llama 4 Maverick | meta/llama-4-maverick via bedrock | 2026-11-01 13:05 | yes | 5 | 3 s |
| Negative framing | Qwen 3.7 Flash | alibaba/qwen3.7-flash via alibaba | 2026-11-01 13:56 | yes | 20 | 51 s |
| Negative framing | Kimi K2 | moonshotai/kimi-k2 via novita | 2026-11-01 09:32 | yes | 25 | 34 s |
| Negative framing | GLM 4.7 FlashX | zai/glm-4.7-flashx via zai | 2026-11-01 16:32 | yes | 25 | 55 s |
| Negative framing | MiniMax M2.5 | minimax/minimax-m2.5 via minimax | 2026-11-01 13:59 | yes | 5 | 23 s |
| Negative framing | GPT-6 Luna | gpt-6-luna | 2026-11-01 08:56 | yes | 2 | 19 s |
| Negative framing | Muse Glimmer 30B | meta/muse-glimmer-30b via togetherai | 2026-11-01 17:45 | yes | 22 | 39 s |
Normalization in this category
Every judgment call made between the raw labels and the numbers above, listed so it is visible and reversible.
ShowHide
Category-scoped readings
Azure OpenAI read as Azure OpenAI Service
Braintrust read as Braintrust Gateway
Cloudflare read as Cloudflare AI Gateway
Datadog read as Datadog LLM Observability
Kong read as Kong AI Gateway
LLM Gateway.io read as LLM Gateway
Portkey.ai read as Portkey
TrueFoundry read as TrueFoundry AI Gateway
Unresolved, counted raw
APIClaw
Anthropic's Workbench
Apigee AI
FastChat
Free LLM Gateway
Google Vertex AI routing
Hugging Face Inference API
Inworld
LlamaStack
Martian Gateway
OneAPI
OpenLLM
Orq.ai AI Router / AI Gateway
Ramp Router
Scale AI
Snorkel
Discontinued, still offered
No shut-down product was recommended here.