BitdepthIndependent grades for inference endpoints
Charts · endpoints over time

Inference endpoints, in charts.

Every chart is drawn from the nightly probe. Free to view, free to cite, free to share. Each one exports clean for X, a deck or a memo.

Finding of the night · Nov 11

On DeepSeek V4 Pro, 4 of 16 endpoints sit outside the reference: gmicloud/fp8, novita/fp8, crusoe/fp8, cloudflare. Ranked endpoints run 50 to 65% on string match.

What the board found

Same model name, different answers.

DeepSeek V4 Pro: outside the reference

On DeepSeek V4 Pro, 4 of 16 endpoints sit outside the reference: gmicloud/fp8, novita/fp8, crusoe/fp8, cloudflare. Ranked endpoints run 50 to 65% on string match.

gmicloud/fp827.9%still listed
novita/fp829.5%still listed
crusoe/fp835.8%not listed since Nov 3
cloudflare21.2%not listed since Nov 3

String match, mean over every night listed.

Cite: Bitdepth · bitdepth.dev/m/deepseek-v4-pro · as of Nov 11, 2026

Full table →

venice/int4, night by night

On gpt-oss-120b, venice/int4 matched the reference on 95% of prompts over its first three nights and 83% over its last three. It now sits outside the reference.

Exact match with the reference each night: venice/int4 against the median of the other endpoints serving gpt-oss-120b.60708090100Oct 27Nov 3Nov 10median of the other 28 endpointsmedian of the other 28 endpointsmedian of the other 28 endpointsvenice/int4venice/int4venice/int4

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

DeepSeek V4 Pro

Price against agreement

Blended price against agreement with the reference for 13 endpoints serving DeepSeek V4 Pro.15%25%35%45%55%65%$0$0.5$1$1.5$2$2.5Blended price per million tokensalibaba/fp8/us-east-1: $1.95 blended, 58.8%alibaba/fp8: $1.77 blended, 58.7%streamlake/fp8: $1.11 blended, 62.2%streamlakeatlas-cloud/fp4: $2.10 blended, 55.7%deepinfra/fp8: $1.62 blended, 62.4%relace/fp4: $1.05 blended, 64.7%relacevenice: $2.06 blended, 60.7%siliconflow/fp8: $1.12 blended, 59.7%siliconflowbaidu/fp8: $1.11 blended, 50.2%digitalocean: $1.31 blended, 51.4%azure/us: $2.39 blended, 55.1%gmicloud/fp8: $1.20 blended, 27.9%gmicloudnovita/fp8: $2.00 blended, 29.5%novita

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

Agreement night by night

Exact match with the reference each night for 16 endpoints serving DeepSeek V4 Pro.30507090Oct 27Nov 4Nov 11alibaba/fp8/us-east-1alibaba/fp8/us-east-1alibaba/fp8alibaba/fp8streamlake/fp8streamlake/fp8atlas-cloud/fp4atlas-cloud/fp4deepinfra/fp8deepinfra/fp8relace/fp4relace/fp4venicevenicesiliconflow/fp8siliconflow/fp8baidu/fp8baidu/fp8digitaloceandigitaloceanazure/usazure/usnebius/fp8crusoe/fp8gmicloud/fp8gmicloud/fp8novita/fp8novita/fp8cloudflare×

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

gpt-oss-120b

Price against agreement

Blended price against agreement with the reference for 25 endpoints serving gpt-oss-120b.75%85%95%$0$0.1$0.2$0.3$0.4$0.5Blended price per million tokensbaseten/fp4: $0.20 blended, 98.0%cerebras/fp16: $0.45 blended, 97.9%crusoe/bf16: $0.10 blended, 97.8%baseten/fp4/us: $0.20 blended, 97.4%deepinfra/fp8: $0.39 blended, 97.4%together: $0.26 blended, 97.2%sambanova: $0.34 blended, 97.1%deepinfra/bf16: $0.070 blended, 96.8%groq: $0.26 blended, 96.8%dekallm/bf16: $0.068 blended, 96.4%parasail/fast: $0.30 blended, 96.4%amazon-bedrock/eu-west-1: $0.26 blended, 96.3%parasail/fp4: $0.26 blended, 96.3%amazon-bedrock: $0.26 blended, 96.1%near-ai/fp4: $0.060 blended, 96.1%near-aisail-research/fp4: $0.15 blended, 96.1%primeintellect/fp4: $0.054 blended, 96.0%primeintellectnebius/fp4: $0.26 blended, 95.8%mancer/fp8: $0.10 blended, 95.7%siliconflow/fp8: $0.26 blended, 95.4%coreweave/fp4: $0.065 blended, 94.9%coreweaveakashml/fp4: $0.069 blended, 94.8%google-vertex/global: $0.16 blended, 94.2%digitalocean: $0.15 blended, 92.1%venice/int4: $0.060 blended, 85.8%venice

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

Agreement night by night

Exact match with the reference each night for 29 endpoints serving gpt-oss-120b.60708090100Oct 27Nov 4Nov 11baseten/fp4baseten/fp4cerebras/fp16cerebras/fp16crusoe/bf16crusoe/bf16baseten/fp4/usbaseten/fp4/usdeepinfra/fp8deepinfra/fp8togethertogethermarasambanovasambanovadeepinfra/bf16deepinfra/bf16groqgroqdekallm/bf16dekallm/bf16amazon-bedrock/eu-west-1amazon-bedrock/eu-west-1parasail/fp4parasail/fp4amazon-bedrockamazon-bedrocknear-ai/fp4near-ai/fp4sail-research/fp4sail-research/fp4primeintellect/fp4primeintellect/fp4akashml/bf16nebius/fp4nebius/fp4mancer/fp8mancer/fp8siliconflow/fp8siliconflow/fp8coreweave/fp4coreweave/fp4google-vertex/globalgoogle-vertex/globalphalanovita/fp4digitaloceandigitaloceanparasail/fastparasail/fastakashml/fp4akashml/fp4venice/int4venice/int4×

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

DeepSeek V4.1 Flash

Price against agreement

Blended price against agreement with the reference for 46 endpoints serving DeepSeek V4.1 Flash.65%75%85%95%$0$0.25$0.5$0.75$1Blended price per million tokenscohere/fp8: $0.24 blended, 90.0%cloudflare: $0.23 blended, 88.6%primeintellect: $0.21 blended, 88.2%makora/fp8: $0.41 blended, 88.0%morph/fp8: $0.12 blended, 87.8%venice/fp8: $0.52 blended, 87.4%coreweave/fp8: $0.52 blended, 87.0%deepinfra/fp8/priority: $0.22 blended, 87.0%wafer/mxfp4: $0.17 blended, 87.0%particle-ai/us: $0.17 blended, 86.9%wafer/us: $0.15 blended, 86.8%particle-ai/us/flex: $0.085 blended, 86.6%primalabs: $0.42 blended, 86.5%stepfun/fp8: $0.24 blended, 86.5%baseten/fast: $1.05 blended, 86.4%fireworks: $0.52 blended, 86.4%atlas-cloud/fp8: $0.34 blended, 86.3%deepinfra/fp8/flex: $0.12 blended, 86.2%modal: $0.44 blended, 86.2%deepinfra/fp8: $0.15 blended, 86.1%relace/fp4: $0.13 blended, 85.9%nebius/fp8: $0.52 blended, 85.8%fireworks/spot: $0.37 blended, 85.7%novita/fp8: $0.29 blended, 85.6%dekallm: $0.31 blended, 85.5%parasail/fp8: $0.52 blended, 85.5%baseten: $0.52 blended, 85.4%mistral/fp8: $0.52 blended, 85.4%alibaba: $0.18 blended, 85.3%inference-net/fp8: $0.21 blended, 85.3%sail-research/fp4/us: $0.11 blended, 85.2%siliconflow/fp8: $0.36 blended, 85.2%fireworks/us: $0.79 blended, 84.9%inference-net/us: $0.21 blended, 84.6%together: $0.52 blended, 84.4%crusoe/fp8: $0.52 blended, 84.1%sail-research/fp4: $0.17 blended, 83.5%gmicloud/fp8: $0.32 blended, 83.0%streamlake/fp8: $0.19 blended, 79.4%ionstream/flex: $0.12 blended, 79.0%ionstream: $0.25 blended, 78.5%digitalocean: $0.32 blended, 76.8%baidu/fp8: $0.26 blended, 76.1%reka: $0.40 blended, 74.9%near-ai: $0.52 blended, 74.8%open-inference/fp4: $0.25 blended, 73.9%

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

Agreement night by night

Exact match with the reference each night for 54 endpoints serving DeepSeek V4.1 Flash.60708090100Nov 4Nov 8Nov 11cohere/fp8cohere/fp8fireworks/ultrafastcloudflarecloudflareprimeintellectprimeintellectmakora/fp8makora/fp8morph/fp8morph/fp8coherevenice/fp8venice/fp8coreweave/fp8coreweave/fp8deepinfra/fp8/prioritydeepinfra/fp8/prioritywafer/mxfp4wafer/mxfp4particle-ai/usparticle-ai/uswafer/uswafer/usparticle-ai/us/flexparticle-ai/us/flexprimalabsprimalabsstepfun/fp8stepfun/fp8baseten/fastbaseten/fastfireworksfireworksatlas-cloud/fp8atlas-cloud/fp8deepinfra/fp8/flexdeepinfra/fp8/flexmodalmodaldeepinfra/fp8deepinfra/fp8relace/fp4relace/fp4nebius/fp8nebius/fp8fireworks/spotfireworks/spotnovita/fp8novita/fp8relace/fastdekallmdekallmparasail/fp8parasail/fp8basetenbasetenmistral/fp8mistral/fp8alibabaalibabainference-net/fp8inference-net/fp8decart/fp8decart/fp8sail-research/fp4/ussail-research/fp4/ussiliconflow/fp8siliconflow/fp8fireworks/usfireworks/usinference-net/usinference-net/usmodelrun/fp4modelrun/fp4togethertogetherphalaphalacrusoe/fp8crusoe/fp8sail-research/fp4sail-research/fp4gmicloud/fp8gmicloud/fp8streamlake/fp8streamlake/fp8ionstream/flexionstreamionstreamdigitaloceandigitaloceandecart/fp4baidu/fp8baidu/fp8rekarekanear-ainear-aiopen-inference/fp4open-inference/fp4××××××

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

Mistral Large 4

Price against agreement

Blended price against agreement with the reference for 2 endpoints serving Mistral Large 4.95%$0$0.25$0.5$0.75$1$1.25Blended price per million tokensmistral/eu: $1.16 blended, 100.0%mistralmistral/zdr: $1.05 blended, 100.0%mistral

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

Agreement night by night

Exact match with the reference each night for 2 endpoints serving Mistral Large 4.90100Nov 5Nov 8Nov 11mistral/eumistral/eumistral/zdrmistral/zdr×

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

GLM 5.3 Flash

Price against agreement

Blended price against agreement with the reference for 38 endpoints serving GLM 5.3 Flash.55%65%75%85%95%$0$0.05$0.1$0.15$0.2$0.25$0.3Blended price per million tokensdarkbloom/fp4: $0.094 blended, 87.5%deepinfra/fp4: $0.24 blended, 91.7%inceptron/fp8: $0.26 blended, 79.2%near-ai/fp8: $0.18 blended, 83.3%nextbit/fp4: $0.20 blended, 83.3%venice: $0.24 blended, 91.7%relace/fp4: $0.050 blended, 85.0%cohere/nvfp4: $0.24 blended, 87.5%coreweave/nvfp4: $0.18 blended, 83.3%crusoe/fp4: $0.24 blended, 83.3%gmicloud/fp8: $0.12 blended, 79.2%nebius/fp4: $0.24 blended, 79.2%streamlake/fp8: $0.14 blended, 87.5%parasail/fast: $0.30 blended, 90.9%friendli: $0.24 blended, 79.2%open-inference/fp4: $0.12 blended, 79.2%primeintellect: $0.10 blended, 87.5%mistral: $0.24 blended, 79.2%morph/fp8: $0.10 blended, 95.7%novita/fp8: $0.13 blended, 87.5%particle-ai/nvfp4: $0.21 blended, 87.5%sail-research/fp4: $0.29 blended, 83.3%inference-net/fp4: $0.17 blended, 79.2%sail-research/us: $0.29 blended, 83.3%together: $0.24 blended, 79.2%z-ai/fp8: $0.24 blended, 91.7%baseten/fp8: $0.24 blended, 79.2%fireworks: $0.24 blended, 66.7%makora/fp4: $0.17 blended, 75.0%wafer/us: $0.077 blended, 79.2%modal/fp4: $0.24 blended, 83.3%siliconflow/fp8: $0.24 blended, 83.3%tenstorrent/fp8: $0.12 blended, 68.2%atlas-cloud/fp8: $0.24 blended, 70.8%digitalocean: $0.18 blended, 83.3%inference-net/us: $0.17 blended, 87.5%wafer/nvfp4: $0.11 blended, 75.0%reka: $0.16 blended, 79.2%

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

Agreement night by night

Exact match with the reference each night for 44 endpoints serving GLM 5.3 Flash.708090100Nov 10Nov 11darkbloom/fp4deepinfra/fp4inceptron/fp8near-ai/fp8nextbit/fp4venicerelace/fp4cohere/nvfp4coreweave/nvfp4crusoe/fp4gmicloud/fp8nebius/fp4riverstreamlake/fp8parasail/fastfriendliopen-inference/fp4primeintellectmistralmorph/fp8novita/fp8particle-ai/nvfp4sail-research/fp4inference-net/fp4sail-research/ustogetherz-ai/fp8baseten/fp8fireworksmakora/fp4wafer/usmodal/fp4phala/nvfp4siliconflow/fp8tenstorrent/fp8atlas-cloud/fp8digitaloceaninference-net/uswafer/nvfp4reka

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

Gemma 4 31B

Price against agreement

Blended price against agreement with the reference for 17 endpoints serving Gemma 4 31B.45%55%65%$0$0.25$0.5$0.75Blended price per million tokenscrusoe/bf16: $0.21 blended, 54.2%novita/bf16: $0.21 blended, 58.3%sambanova: $0.57 blended, 54.5%siliconflow/fp8: $0.81 blended, 56.5%deepinfra/fp8: $0.25 blended, 54.2%parasail/fp8: $0.21 blended, 54.2%deepinfra/fp8/flex: $0.20 blended, 54.2%deepinfra/fp8/priority: $0.38 blended, 54.2%friendli: $0.21 blended, 62.5%coreweave/fp4: $0.16 blended, 54.2%modal: $0.28 blended, 54.2%deepinfra/turbo: $0.23 blended, 54.2%dekallm: $0.21 blended, 59.1%chutes/fp4: $0.18 blended, 54.2%nextbit/fp4: $0.21 blended, 54.2%venice/fp4: $0.18 blended, 54.2%primeintellect/fp4: $0.23 blended, 54.2%

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

Agreement night by night

Exact match with the reference each night for 20 endpoints serving Gemma 4 31B.708090100Nov 10Nov 11crusoe/bf16novita/bf16sambanovasiliconflow/fp8deepinfra/fp8parasail/fp8deepinfra/fp8/flexdeepinfra/fp8/priorityfriendlicoreweave/fp4io-netmodaldeepinfra/turbodekallmchutes/fp4modelrun/fp4nextbit/fp4venice/fp4primeintellect/fp4

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

MiMo V2.6 Pro

Price against agreement

Blended price against agreement with the reference for 5 endpoints serving MiMo V2.6 Pro.45%55%65%75%$0$1$2$3$4$5Blended price per million tokensstreamlake: $0.65 blended, 62.5%streamlakedeepinfra/fp8: $0.49 blended, 66.7%deepinfragmicloud/bf16: $0.41 blended, 66.7%gmicloudprimalabs/fp8: $5.00 blended, 58.3%primalabsxiaomi/fp8: $0.54 blended, 70.8%xiaomi

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

Agreement night by night

Exact match with the reference each night for 5 endpoints serving MiMo V2.6 Pro.8090100Nov 10Nov 11streamlakedeepinfra/fp8gmicloud/bf16primalabs/fp8xiaomi/fp8

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

All models

Declared quantization against measured state

Matches referenceFirst lookOutside reference
Endpoints by declared quantization, split by measured state, all models.unknownunknown: 12 matches reference12unknown: 43 first look43unknown: 1 outside reference56fp8fp8: 9 matches reference9fp8: 43 first look43fp8: 3 outside reference355fp4fp4: 11 matches reference11fp4: 31 first look3142bf16bf16: 4 matches reference4bf16: 3 first look37nvfp4nvfp4: 6 first look66mxfp4mxfp4: 2 first look2fp16fp16: 1 matches reference1int4int4: 1 outside reference1

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

Moved this period

Mean exact match of the last three nights against the first three.

  • venice/int4 on gpt-oss-120b: -11.4 points

Cite: Bitdepth · bitdepth.dev/charts · as of Nov 11, 2026

Glossary

Speak the language.

Endpoint

One provider serving one model at one declared precision on OpenRouter.

Declared quantization

The precision the provider lists: bf16 or fp16 is full precision; fp8, fp4 and int4 are smaller and cheaper to run.

Reference

The answer the full-precision endpoints give. What every endpoint is graded against.

Agreement

The share of prompts where an endpoint gives the reference answer.

Quality-adjusted price

Blended price divided by agreement. What a correct answer costs.

Drift

An endpoint whose answers move over the period, measured against its own first nights.

Numbers against a stated reference. Not a claim about intent.Listed prices from OpenRouter, read nightly; grades are Bitdepth's.