KI-Verzeichnis

Claude Haiku 4.5

Anthropic

Empfohlen

Anthropics Preis-Leistungs-Basis für $1/$5 per 1M tokens — die schnelle, günstige Stufe, an der sich die Konkurrenz misst, mit 200K Kontext und 64K Output. Am besten für latenzkritische Arbeit mit hohem Volumen: Klassifikation, Extraktion, Zusammenfassung, Routing. Greifen Sie zu Opus 4.8 oder Sonnet 5, wenn tiefes mehrstufiges Reasoning gefragt ist.

Eingabe

$1 / 1M tokens

Ausgabe

$5 / 1M tokens

Kontext

200K ctx; legacy, Haiku 5.5 is current

Claude Opus 4.8

Anthropic

Empfohlen

Claude Opus 4.8 is Anthropic's previous-generation Opus, now listed as legacy: 1M-token context, 128K output, $5/$25 per 1M tokens, retirement not before 28 May 2027. It remains our pick for review-heavy work such as code review and constrained refactoring, where deliberate pacing beats persistence. New workloads should start on Claude Opus 5.5 at $4/$20, which Anthropic now recommends.

Eingabe

$5 / 1M tokens

Ausgabe

$25 / 1M tokens

Kontext

1M tokens

Claude Opus 5

Anthropic

Empfohlen

Claude Opus 5 delivers near-Fable 5 intelligence at $5/$25 per 1M tokens, with a 1M-token context and 128K output. Anthropic now lists it as legacy: still available, retirement not before 24 July 2027, and Claude Opus 5.5 ($4/$20) is its recommended starting point. Recommended for teams already running Opus 5; new workloads should start on Opus 5.5, which is cheaper per token.

Eingabe

$5.00 / 1M tokens

Ausgabe

$25.00 / 1M tokens

Kontext

1M ctx; 128K out; legacy; fast $10/$50

Claude Sonnet 5

Anthropic

Empfohlen

Anthropic's most agentic Sonnet at launch delivers performance close to Opus 4.8 at 40 percent of the Opus rate. The $2/$10 per 1M token price is the standard rate: Anthropic cancelled the increase to $3/$15 that was scheduled for 1 September 2026. The newer tokenizer produces about 30 percent more tokens for the same text, so recount budgets before migrating from Sonnet 4.6. Anthropic now lists Sonnet 5 as legacy; Claude Sonnet 5.5, at the same $2/$10, is the current Sonnet, so start new work there.

Eingabe

$2 / 1M

Ausgabe

$10 / 1M

Kontext

1M tokens; batch $1/$5; cache read $0.20

DeepSeek V4 Flash

DeepSeek

Empfohlen

DeepSeek retired V4 Flash from its API on 10 September 2026: the deepseek-v4-flash id is still accepted, but its requests are served by V4.1-Flash and billed at the Flash rate of $0.15 input and $0.60 output per 1M tokens off-peak. The MIT-licensed weights (284B parameters, 13B active, 1M-token context) remain on Hugging Face, so this is now a frozen checkpoint for self-hosting and reproducible evals. For new API work, call deepseek-flash.

Eingabe

$0.15 / 1M tokens

Ausgabe

$0.60 / 1M tokens

Kontext

Retired Sep 10; id serves V4.1-Flash. Peak 2x M-F

DeepSeek V4 Pro

DeepSeek

Empfohlen

Strong price-performance on open MIT weights; the 2x peak window is 01:00-04:00 and 06:00-10:00 UTC on weekdays only, so weekends bill entirely off-peak.

Eingabe

$0.66 / 1M tokens

Ausgabe

$1.98 / 1M tokens

Kontext

1M tokens; 2x price at weekday peak hours

Gemini 2.5 Pro

Google

Empfohlen

Jetzt die Budget-Empfehlung für lange Kontexte: Eine Preissenkung um 40 % (Juni 2026) macht vollständige Korpus-Durchläufe auch für mittlere Budgets erschwinglich. Es bietet das größte verfügbare Kontextfenster mit starker Google-Ökosystem-Integration.

Eingabe

$1.25 / 1M tokens

Ausgabe

$10 / 1M tokens

Kontext

1M tokens

Gemini 3.5 Flash

Google

Empfohlen

Googles Kosteneffizienz-Highlight für agentische Desktop-Automatisierung zu $1.50/$9 per 1M tokens mit ~1M Kontext. Natives Computer Use erreicht 78,4 auf OSWorld-Verified — nur 0,3 Punkte hinter GPT-5.5, zu rund einem Drittel der Kosten. DER relevante Vorbehalt: Das Computer-Use-Tool ist eine öffentliche Preview und noch nicht in der Produktions-API — Endpoint-Verfügbarkeit vor der Anbindung prüfen.

Eingabe

$1.50 / 1M tokens

Ausgabe

$9 / 1M tokens

Kontext

1M tokens

Gemini 3.5 Flash-Lite

Google

Empfohlen

Gemini 3.5 Flash-Lite ist Googles schnellstes und günstigstes Modell der 3.5-Serie mit 350 Tok/s und $0,30/$2,50. Der Generationssprung von 3.1 Flash-Lite ist die eigentliche Geschichte — es übertrifft seinen Vorgänger dramatisch bei wichtigen agentischen Benchmarks und schlägt das vollwertige Gemini 3 Flash bei Coding- und Computer-Use-Aufgaben. Für Entwickler, die hochdurchsatzstarke agentische Subagents, Such-Pipelines und Dokumentenverarbeitung skalieren, ist Flash-Lites Preis-Leistungs-Verhältnis in Googles Produktpalette unerreicht.

Eingabe

$0.30 / 1M

Ausgabe

$2.50 / 1M

Kontext

1M context, 64K output, 350 tok/s

Gemini 3.6 Flash

Google

Empfohlen

Gemini 3.6 Flash is Google's new default Flash-tier workhorse: better than 3.5 Flash on every benchmark and cheaper, at $0.75/$3.75 per 1M tokens through 2026-12-31, rising to $1.50/$7.50 from 2027-01-01. The 17% token-efficiency gain lowers per-task cost, and built-in Computer Use via the API makes it a credible agentic coding option. For Google-ecosystem developers it is the clear upgrade; budget for the January rate step.

Eingabe

$0.75 / 1M (through 2026-12-31)

Ausgabe

$3.75 / 1M (through 2026-12-31)

Kontext

1M context, 64K output; $1.50/$7.50 from 2027-01-01

Gemini 3.7 Flash

Google

Empfohlen

Gemini 3.7 Flash ist Googles bislang stärkstes Flash-Arbeitspferd für Coding und Agents — ein echter Schritt nach vorn gegenüber 3.6 Flash, bis 2026 zum halben vorherigen Preis. Bei den härtesten Terminal- und Computer-Use-Aufgaben liegt es weiterhin hinter der Frontier. Ein klares Upgrade für kostenbewusste Agent-Builder und solche im Google-Ökosystem; kein Frontier-Code-Ersatz.

Eingabe

$0.75 / 1M

Ausgabe

$3.75 / 1M

Kontext

1M context, 64K output

Gemini 3.8 Flash

Google

Empfohlen

Gemini 3.8 Flash is Google's strongest Flash workhorse yet — a genuine capability upgrade over the recommended 3.7 Flash at the same introductory price, with real gains across software engineering, agents and professional domains. The deciding caveat is Google's own: 3.8 "works harder," so per-task token burn runs well above 3.7, which remains the efficiency-first pick. Adopt 3.8 for long-horizon coding, agents and finance or legal work — benchmark your own workload before the intro price expires.

Eingabe

$0.75 / 1M

Ausgabe

$3.75 / 1M

Kontext

1M context, 64K output

Gemini 3.8 Live Extended Thinking

Google

Empfohlen

Gemini 3.8 Live Extended Thinking is the voice-agent model to start with: it tops Artificial Analysis' Speech to Speech Index and reasons and calls tools in the background while it keeps talking, at the same rate as the default Gemini 3.8 Live. The caveat is the Live API surface: no structured outputs, caching, code execution or batch. Right for teams building real-time voice agents that must solve multi-step problems mid-call.

Eingabe

$0.75 / 1M (text)

Ausgabe

$4.50 / 1M (text)

Kontext

131K in, 65K out; audio $3/$12 per 1M

GPT-4.1

OpenAI

Empfohlen

Das Standard-Migrationsziel für GPT-4-Turbo-Workloads: gleiche Anfragestruktur, niedrigere Kosten, 1M-Token-Kontext – wandeln Sie veraltete function_call-Payloads in das Tools-Format um und machen Sie weiter.

Eingabe

$2 / 1M tokens

Ausgabe

$8 / 1M tokens

Kontext

1M tokens

GPT-5.5

OpenAI

Empfohlen

GPT-5.5 is OpenAI's April 2026 general-purpose model at $5 input and $30 output per 1M tokens, with a 1M+ token context (922K in, 128K out). It is no longer OpenAI's only deployable frontier option: the GPT-5.6 family is on the public price list, and GPT-6 Sol lists at $2/$10. It stays recommended as a stable, widely supported production baseline, but new workloads should test GPT-6 Sol or GPT-5.6 Terra first.

Eingabe

$5 / 1M tokens

Ausgabe

$30 / 1M tokens

Kontext

1M+ tokens (922K in / 128K out)

GPT-5.6 Sol

OpenAI

Empfohlen

GPT-5.6 Sol is no longer OpenAI's best-value frontier model: its successor GPT-6 Sol, released 22 September 2026, lists at $2/$10 per 1M tokens against Sol's $4/$20. It stays recommended because its coding lead has third-party backing, while every GPT-6 Sol score so far is OpenAI's own. OpenAI lists no deprecation, and the $4/$20 rate is promotional through at least 21 November 2026. Test GPT-6 Sol before committing new workloads.

Eingabe

$4 / 1M tokens

Ausgabe

$20 / 1M tokens

Kontext

Short $4/$20 · long $8/$30 per 1M · promo

GPT-Live-Transcribe

OpenAI

Empfohlen

OpenAIs Streaming-Sprache-zu-Text-Modell mit niedriger Latenz liefert Live-Transkript-Deltas zu 0,017 $/Min. mit einstellbarer Latenz und kontextbewusster Transkription. Es verbessert die Fehlerrate von gpt-realtime-whisper und unterstützt Echtzeit-Keyword-Hinweise, Sprachhinweise und Konfigurationsaktualisierungen während der Sitzung über WebSocket oder WebRTC. Der Kompromiss: etwa das 3,8-fache der Kosten von Batch-GPT-Transcribe und keine Diarisierung, Wort-Timestamps oder Konfidenzwerte. Am besten für Live-Untertitelung, Meeting-Assistenten und Sprachagenten, bei denen die Anzeige von Transkripten unter einer Sekunde wichtiger ist als die Kosten pro Minute.

Eingabe

$0.017 / minute

Ausgabe

N/A

Kontext

Per-minute realtime audio billing

GPT-Transcribe

OpenAI

Empfohlen

OpenAIs neuestes Batch-Speech-to-Text-Modell erreicht 3,31 % AA-WER bei 0,0045 $/Min. — 25 % günstiger und messbar genauer als sein gpt-4o-transcribe-Vorgänger. Es unterstützt Kontext-Prompting, Keyword-Hinweise und Spracherkennung in über 22 Sprachen und ist damit der beste Standard für asynchrone Dateitranskription auf der OpenAI-Plattform. Einschränkung: keine Diarisierung oder wortgenaue Zeitstempel; nutzen Sie gpt-4o-transcribe-diarize für Sprecherlabels oder gpt-live-transcribe für Echtzeitanforderungen. Am besten für Entwickler, die Meeting-Transkription, Medienverarbeitung und Sprachschnittstellen auf der OpenAI API aufbauen.

Eingabe

$0.0045 / minute

Ausgabe

N/A

Kontext

Per-minute audio billing

Claude Fable 5

Anthropic

Bedingt

Anthropic's Mythos-class frontier model, the strongest publicly available Claude at its June 2026 launch, now listed as legacy behind Claude Fable 5.1. The verdict stays conditional on two constraints: API pricing of $10/$50 per 1M tokens, and a safety classifier that over-triggered on routine coding tasks at launch, routing them to an Opus 4.8 fallback. Subscribers reach Fable through Claude Max at 50% of weekly limits, or through usage credits on Pro. Anthropic's current lineup lists Fable 5.1 and Opus 5.5 for new work.

Eingabe

$10 / 1M tokens

Ausgabe

$50 / 1M tokens

Kontext

1M ctx, 128K out; legacy; Max plans at 50% limits

Claude Fable 5.1

Anthropic

Bedingt

Anthropic's new frontier flagship is the most capable model publicly available, and it largely fixes the safeguard over-triggering that made Fable 5 frustrating for routine coding. The verdict stays conditional on cost: independent measurement shows more spend per task than Fable 5 despite the cheaper cache reads, because the model thinks longer. Reach for it on demanding reasoning and long-horizon agentic work. For everyday coding, Claude Opus 5.5 at $4/$20 per 1M is Anthropic's recommended starting model and the better default.

Eingabe

$10 / 1M tokens

Ausgabe

$50 / 1M tokens

Kontext

1M context, 128K output; cache $0.25/1M

Claude Haiku 5.5

Anthropic

Bedingt

Claude Haiku 5.5 is a conditional pick: Anthropic's fastest and cheapest current Claude at $0.10 input and $0.50 output per 1M tokens, with a 1M context and adjustable effort. Prompts over 100K tokens cost five times as much, and every benchmark is Anthropic's own. Right for high-volume classification, extraction, routing and subagent work. Use Sonnet 5.5 for complex agentic coding, where Anthropic's own Terminal-Bench 4.0 numbers put it well ahead.

Eingabe

$0.10 / 1M tokens

Ausgabe

$0.50 / 1M tokens

Kontext

1M ctx; prompts over 100K cost $0.50/$2.50

Claude Mythos 5.1

Anthropic

Bedingt

Same weights as Claude Fable 5.1 at the same $10/$50 rate, with cyber and life-sciences safeguards loosened for vetted users. The catch is access: Mythos 5.1 ships only through Anthropic's trusted-access programs (Cyber Verification and Life Sciences Verification), so there is no self-serve API and no way to buy your way in. Conditional for the same reason as Mythos 5: elite capability behind a vetting gate. Everyone else should use Fable 5.1.

Eingabe

$10 / 1M tokens

Ausgabe

$50 / 1M tokens

Kontext

1M context, $0.25/1M cache; vetted access

Claude Opus 5.5

Anthropic

Bedingt

Claude Opus 5.5 is the model Anthropic now recommends starting with for most workloads, built for long-running agentic coding and knowledge work at rates 20% below Claude Opus 5. Conditional because every benchmark is Anthropic's own and migration is not drop-in: thinking cannot be turned off, forced tool use returns an error, and most cybersecurity tasks are rerouted to Opus 4.8. Right for Claude Opus 5 teams that can absorb those breaking changes.

Eingabe

$4 / 1M tokens

Ausgabe

$20 / 1M tokens

Kontext

1M ctx; 128K out; cache $0.20; fast $8/$40

Claude Sonnet 4.6

Anthropic

Bedingt

Superseded by Claude Sonnet 5 (shipped Jun 30, 2026), which Anthropic now lists as legacy behind Claude Sonnet 5.5 as the current Sonnet. Both list at $2/$10 per 1M tokens against Sonnet 4.6's $3/$15, since Anthropic cancelled the scheduled rise to $3/$15. Sonnet 5 also adds a 13.4-point Terminal-Bench 2.1 jump. Choose 4.6 only for pinned deployments or where Sonnet 5's ~30% higher token count and removed temperature/top_p/top_k are disqualifying.

Eingabe

$3 / 1M tokens

Ausgabe

$15 / 1M tokens

Kontext

1M tokens

Claude Sonnet 5.5

Anthropic

Bedingt

Conditional. Claude Sonnet 5.5 is Anthropic's default Sonnet model on the Anthropic API, on the same rate card as Sonnet 5, and Anthropic says it runs faster and costs less per task. Every benchmark is Anthropic's own, and five breaking changes hit code written for Sonnet 5. Right for teams ready to re-test agentic coding and knowledge work; hold off if you rely on forced tool choice or temperature control.

Eingabe

$2 / 1M tokens

Ausgabe

$10 / 1M tokens

Kontext

1M ctx; batch $1/$5; cache read $0.10

DeepSeek V4.1 Flash

DeepSeek

Bedingt

DeepSeek V4.1 Flash, the model now behind DeepSeek's deepseek-flash API, is a strong-value open-weight pick: MIT-licensed weights, image input, a 1M-token context and an Artificial Analysis score well above average for comparable models. Conditional because its standout coding-agent results are DeepSeek's own, Artificial Analysis measured it as very verbose, and peak hours double the rate. Right for cost-sensitive pipelines that can run off-peak or self-host; wrong for buyers who need independently verified agent performance before committing.

Eingabe

$0.15 / 1M tokens

Ausgabe

$0.60 / 1M tokens

Kontext

1M tokens; 2x price at weekday peak hours

Conditional. Eleven v4 is ElevenLabs' long-form text-to-speech model (id eleven_v4), covering 90+ languages with multi-speaker dialogue and voice cloning, and accepting up to 10,000 characters per request. The deciding caveat is price: $0.022 per 1K characters is a 72%-off promo that ends Oct 12 against a $0.08 list rate, and the quality claims are ElevenLabs' own, untested by us. Suits teams producing content and long-form audio who budget at list price.

Eingabe

$0.08 / 1K characters

Ausgabe

Billed per character (single rate)

Kontext

List $0.08 / 1K chars; 10K char limit

Conditional. Eleven v4 Turbo (id eleven_v4_turbo) is ElevenLabs' real-time speech model, stating a median inference latency of about 100 ms, excluding application and network latency, across 90+ languages via a Text to Dialogue WebSocket. The deciding caveat is that the latency figure is vendor-stated and untested by us, and $0.011 per 1K characters is a promo ending Oct 12 against a $0.04 list rate. Suits builders of voice agents and interactive characters.

Eingabe

$0.04 / 1K characters

Ausgabe

Billed per character (single rate)

Kontext

List $0.04 / 1K chars; median latency ~100 ms

FLUX 3

Black Forest Labs

Bedingt

FLUX 3 is Black Forest Labs' unified multimodal model for image, video, audio and robot-action prediction. Its Video component is now pay-as-you-go through BFL's dashboard and API, from $0.06 per second at Draft HD to $0.80 at UHD, with five seconds of Standard HD costing $0.85. Conditional because every preference result is BFL's own, Image and Audio are not yet released, and the open-weight FLUX 3 Dev is still listed as coming soon.

Eingabe

$0.17 / sec (Standard HD text-to-video)

Ausgabe

Included in per-second video rate

Kontext

Video GA pay-as-you-go; $0.06-$0.80/s by res

Gemini 3.1 Pro

Google

Bedingt

Googles leistungsfähigstes öffentlich verfügbares Pro-Tier-Modell mit 1M-Token-Kontext zu wettbewerbsfähigen Preisen. Vorgänger-Generation: Gemini 3.6 und 3.5 Flash schlagen es jetzt bei Coding- und agentischen Benchmarks zu niedrigeren Kosten. Bleib dabei in Google Cloud/Vertex AI; ansonsten liefern die Flash-Reihe oder Wettbewerber mehr.

Eingabe

$2 / 1M

Ausgabe

$12 / 1M

Kontext

1M tokens

Gemini 3.8 Flash Cyber

Google

Bedingt

Gemini 3.8 Flash Cyber is Google's most capable cybersecurity model, a defensive specialist already in production at Chrome Security, Cloud Vulnerability Research and partner Wiz, with frontier vulnerability-discovery and patching results at a fraction of the leaders' rollout cost. The deciding caveat is access: it is gated to vetted defenders through the Fairwind Program, with no public API or published rate card. We rate it conditional rather than caution like the sibling Gemini 3.5 Flash Cyber because 3.8 ships published benchmark figures and named production deployments where 3.5 had neither, even though both remain gated. Right for organizations that can qualify as trusted defenders; everyone else cannot buy it today.

Eingabe

N/A (no public rate card)

Ausgabe

N/A (no public rate card)

Kontext

Fairwind-gated; no public API

Gemini 3.8 Flash TTS

Google

Bedingt

Conditional. Gemini 3.8 Flash TTS is Google's flagship text-to-speech model, available in the Gemini API with voice design, consented voice replication and SynthID watermarking on every clip. The deciding caveat is cost: today's paid rates are launch prices that double when the launch window closes, and its Hume AI leaderboard placements are Google-reported, not measured by us. Suits developers building expressive narration, character voices or dubbing who can budget for the post-launch rate.

Eingabe

$0.50 / 1M (text)

Ausgabe

$9.00 / 1M (audio)

Kontext

Launch rate to 2026-12-31; $1/$18 from 2027

Gemini 3.8 Flash-Lite TTS

Google

Bedingt

Conditional. Gemini 3.8 Flash-Lite TTS is Google's cost tier for high-volume speech: stable in the Gemini API, built into Google Vids, and billed at the same text input rate as Flash TTS with a lower audio output rate. Google ranks it second only to Flash TTS on Hume AI's Overall Quality Index, but that ranking is Google-reported and its launch rates are temporary, doubling once the promotion ends. Suits voice-agent cascades and bulk narration where cost per clip matters more than acting range.

Eingabe

$0.50 / 1M (text)

Ausgabe

$6.00 / 1M (audio)

Kontext

Launch rate to 2026-12-31; $1/$12 from 2027

Gemini 3.8 Live

Google

Bedingt

Gemini 3.8 Live is Google's default Live API model for low-latency voice agents that keep talking while tool calls finish in the background. Conditional because Google bills Gemini 3.8 Live Extended Thinking at the identical rate, and Artificial Analysis scores that sibling higher on its Speech to Speech Index, so you pay the same for less reasoning. Pick 3.8 Live only when turn latency matters more than reasoning depth: developers shipping real-time voice experiences on the Gemini API.

Eingabe

$0.75 / 1M (text)

Ausgabe

$4.50 / 1M (text)

Kontext

131K in, 65K out; audio $3/$12 per 1M

GLM-5.2

Z.ai

Bedingt

Das führende Open-Weight-Modell fürs Coding mit 1M-Token-Kontext und MIT-Lizenz. Es übertrifft GPT-5.5 in mehreren Benchmarks zu einem Bruchteil der Kosten, liegt aber bei den härtesten Langhorizont-Aufgaben weiterhin hinter Claude Opus 4.8.

Eingabe

$1.40 / 1M tokens

Ausgabe

$4.40 / 1M tokens

Kontext

1M tokens

GLM-5.3

Z.ai

Bedingt

GLM-5.3 is Z.ai's open-weights coding and cyber-defense model on the GLM-5.2 base, with every gain from post-training. Pricing is published at $1.40 input / $4.40 output per 1M, the same as GLM-5.2, and the weights are now on Hugging Face. It stays conditional because its headline coding and cyber scores come from Z.ai's own evaluations: reach for defensive security and agentic automation, but audit it yourself before standardizing.

Eingabe

$1.40 / 1M tokens

Ausgabe

$4.40 / 1M tokens

Kontext

1M ctx; Flash $0.15/$0.50, launch promo ended

GPT-4o (Retired from ChatGPT, API-only)

OpenAI

Bedingt

GPT-4o (May 2024) has been superseded by the GPT-5.x family and delisted from OpenAI's flagship API pricing page. It remains available through the API but is no longer a current-generation model. Use only where migration to GPT-5.x is blocked; otherwise, GPT-5.5 or GPT-5.4 offer better performance at comparable cost.

Eingabe

$2.50 / 1M tokens

Ausgabe

$10 / 1M tokens

Kontext

128K tokens

GPT-5.5-Cyber

OpenAI

Bedingt

85,6 % CyberGym — bestes Einzelmodell-Ergebnis — aber Zugang nur für geprüfte Verteidiger.

Eingabe

N/A

Ausgabe

N/A

Kontext

Trusted Access for Cyber; not commercial API

GPT-5.6 Luna

OpenAI

Bedingt

GPT-5.6 Luna is OpenAI's previous-generation budget model at $0.20 input and $1.20 output per 1M tokens short context, generally available through the API and Codex. Its successor GPT-6 Luna, released 22 September 2026, lists at $0.10/$0.50, so Luna is no longer the low-cost default for new OpenAI work. It remains conditional because benchmarks are vendor-reported and it lacks activation classifiers; OpenAI lists no deprecation, so existing pipelines need not migrate yet.

Eingabe

$0.20 / 1M tokens

Ausgabe

$1.20 / 1M tokens

Kontext

Short $0.20/$1.20 · long $0.40/$1.80 per 1M

GPT-5.6 Terra

OpenAI

Bedingt

GPT-5.6 Terra is OpenAI's mid-tier workhorse: GPT-5.5-class capability on OpenAI's own claim, at $2.00 input and $12.00 output per 1M tokens at short context ($4.00 / $18.00 long) after the 20% price cut of 30 July 2026. It is generally available via the API and Codex. Remains conditional: all benchmarks are vendor-reported without independent reproduction, and Terra is not selectable in the standard ChatGPT model picker.

Eingabe

$2.00 / 1M tokens

Ausgabe

$12.00 / 1M tokens

Kontext

Short $2.00/$12.00 · long $4.00/$18.00 per 1M

GPT-5.6-Cyber

OpenAI

Bedingt

GPT-5.6-Cyber ist OpenAIs speziell trainiertes Cybersicherheitsmodell, aufbauend auf GPT-5.6 Sol und ausschließlich über das geschützte Daybreak-Red-Programm verfügbar. Es ist das erste Modell, das OpenAIs 'Hohe'-Cyber-Fähigkeitsschwelle erreicht, mit Exploit-Chain- und Zero-Day-Leistung, die universelle Modelle weit übertrifft.

Eingabe

$12.50 / 1M tokens

Ausgabe

$75 / 1M tokens

Kontext

400K tokens

GPT-6 Astra

OpenAI

Bedingt

Astra ships at $10/$50 per 1M tokens short context, $20/$75 long context, with a 1,050,000-token context window and 128K max output. On OpenAI's own reported benchmarks it leads on computer use, terminal work, long context and cyber. We previously placed it third on the Artificial Analysis Intelligence Index at 61.2; we could not source that figure and have withdrawn it, and Artificial Analysis publishes no Astra score, so we have no vendor-neutral general-intelligence ranking for it. Conditional on two constraints: Fast mode is unavailable under EU data residency, and OpenAI disclosed decreased monitorability versus GPT-5.6 Sol. Route to it by task; do not make it your default.

Eingabe

$10 / 1M tokens

Ausgabe

$50 / 1M tokens

Kontext

1.05M tokens; short-ctx standard rate

GPT-6 Luna

OpenAI

Bedingt

GPT-6 Luna is OpenAI's new budget model and the cheapest way into GPT-6, at half GPT-5.6 Luna's input rate, with a 1.05M-token context and text and image input. Conditional because every published result is OpenAI's own and prompts over 272K input tokens bill at higher rates. GPT-5.6 Luna stays on the price list with no deprecation notice, so migration is optional. Right for teams running high-volume, everyday API work at scale.

Eingabe

$0.10 / 1M tokens

Ausgabe

$0.50 / 1M tokens

Kontext

1.05M ctx; over 272K $0.20/$0.75; cache $0.01

GPT-6 Sol

OpenAI

Bedingt

GPT-6 Sol is the lower-priced companion to GPT-6 Astra and succeeds GPT-5.6 Sol at half its per-token rate, with a 1.05M-token context and 128K output. Conditional because every published score is OpenAI's own and prompts over 272K input tokens bill at higher rates. GPT-5.6 Sol stays on the price list with no deprecation notice, so nothing forces a migration. Right for teams routing hard coding and agent work to OpenAI.

Eingabe

$2 / 1M tokens

Ausgabe

$10 / 1M tokens

Kontext

1.05M ctx; over 272K $4/$15; cache $0.20

GPT-6.1 Sol

OpenAI

Bedingt

Conditional on evidence: GPT-6.1 Sol is OpenAI's model for complex coding, computer use and professional work, which OpenAI itself describes as near-Astra performance at a lower cost. We could not read the launch post, so no benchmark is stored and the quality claim is OpenAI's alone. Prompts past 272K input tokens bill at higher rates, and reasoning cannot be set to none or minimal. Suits teams building agents on the Responses API who will run their own evals first.

Eingabe

$2 / 1M tokens

Ausgabe

$10 / 1M tokens

Kontext

1.05M ctx; over 272K $4/$15; cache $0.10

GPT-Live-1

OpenAI

Bedingt

GPT-Live-1 is OpenAI's full-duplex voice model: it listens and speaks at the same time and hands reasoning and tool use to a backend model. Since 10 September 2026 it is generally available in the API on the Live endpoint at $0.05 per minute, with backend model and tool usage billed separately. Conditional because its benchmarks are OpenAI's own, language quality is uneven, and real cost per call depends on the backend model you pair it with.

Eingabe

$0.05 / minute (voice session)

Ausgabe

Included in per-minute session rate

Kontext

Live API; backend model + tools billed separately

GPT-Live-1 mini

OpenAI

Bedingt

Die kostenlose Variante von GPT-Live-1 — dieselbe Vollduplex-Architektur in einem kleineren, schnelleren Paket, das als Standard-Sprachmodell für ChatGPT Free-Nutzer dient. Teilt die Kerninnovationen: kontinuierliche Interaktion, Wake-Word-Unterstützung und GPT-5.5-Delegation. Dieselben Einschränkungen: keine API, ungleichmäßige Sprachqualität. Eine wesentliche Verbesserung gegenüber dem alten turn-basierten Sprachmodus für kostenlose Nutzer.

Eingabe

Free (ChatGPT)

Ausgabe

Free (ChatGPT)

Kontext

ChatGPT Voice only

Grok 4.5

SpaceXAI

Bedingt

Competitively priced coding and agentic model at $2/$6 per 1M tokens up to 200K tokens of context, doubling to $4/$12 above that, with a 500K window. xAI's published benchmarks place it mid-pack behind Fable 5 and GPT-5.5, though it leads Opus 4.8 on DeepSWE and Terminal-Bench 2.1. The real story is cost efficiency: 4.2x more token-efficient than Opus 4.8 on SWE-Bench Pro. Best for teams needing strong coding at a fraction of frontier pricing; xAI's docs list only US regions.

Eingabe

$2 / 1M tokens

Ausgabe

$6 / 1M tokens

Kontext

500K ctx; 200K+ $4/$12; cache $0.30 (200K+ $0.60)

Grok 4.6

SpaceXAI

Bedingt

Grok 4.6 is SpaceXAI's frontier model for coding and agentic work at $2 input and $6 output per 1M tokens under 200K, with a 500K-token context; prompts of 200K or more bill at $4/$12. At launch it scored 61 on the Artificial Analysis Intelligence Index, tying GPT-5.6 Sol, but that is still one independent evaluator. Grok 4.7 now lists at the same price, so compare the two before committing new work.

Eingabe

$2.00 / 1M tokens

Ausgabe

$6.00 / 1M tokens

Kontext

500K ctx; 200K+ $4/$12; cache $0.50

Grok 4.7

SpaceXAI

Bedingt

Grok 4.7 keeps Grok 4.6's price and speed, and xAI says it works longer on hard tasks and checks its own work more carefully. Conditional because every published score is xAI's own, prompts of 200K tokens or more bill at double the headline rate, and the Batch API is not supported. Right for coding teams already working in Cursor or Grok Build.

Eingabe

$2.00 / 1M tokens

Ausgabe

$6.00 / 1M tokens

Kontext

500K ctx; 200K+ $4/$12; fast variant 2x price

Grok Voice 3.0

SpaceXAI

Bedingt

Grok Voice 3.0 is SpaceXAI's speech-to-speech model for real-time voice agents, priced at $0.08 per minute, the same as Grok Voice 2.0. Artificial Analysis shows the highest Speech-to-Speech Index score, 86.9, with 0.81 seconds to first audio, but xAI's other benchmark and customer figures are its own. Conditional because we have not tested it and the grok-voice-latest alias moved to it automatically. It suits teams building phone or in-app voice agents on a managed model.

Eingabe

$0.08 / min of audio (speech-to-speech)

Ausgabe

Included in per-minute audio rate

Kontext

Voice Agent API; $4.80 / hr

Grok Voice Think Fast 2.0

SpaceXAI

Bedingt

SpaceXAIs Sprach-zu-Sprache-Modell der nächsten Generation mit einer parallelen Reasoning-Architektur, die es deutlich intelligenter macht als herkömmliche Sprachmodelle.

Eingabe

$0.08 / min of audio (speech-to-speech)

Ausgabe

Included in speech-to-speech rate

Kontext

API-only

Grok Voice Transcribe 2.0

SpaceXAI

Bedingt

Grok Voice Transcribe 2.0 is SpaceXAI's speech-to-text model at $0.10 per hour of audio in batch, with speaker labels, word timestamps and key terms included. Conditional because every accuracy figure is SpaceXAI's own: it says the model is twice as accurate as 1.0 and first among 32 streaming models on Artificial Analysis, and we could not check that leaderboard. Right for voice-agent and call-transcription developers who will test it on their own audio; we have not tested it.

Eingabe

$0.10 / hour of audio (batch)

Ausgabe

Included in per-hour audio rate

Kontext

Streaming $0.20 / hour of audio

Hy3

Tencent

Bedingt

Apache 2.0 open-weight MoE agentic model (295B total, 21B active) with strong tool-calling reliability and anti-hallucination behavior, but coding trails GLM-5.2 and self-hosting demands 8+ GPUs. The free OpenRouter tier that ran through Jul 21 is gone; OpenRouter now lists six providers at $0.13-$0.20 input and $0.53-$0.80 output per 1M tokens. Best for self-hosted coding agents and long-context document work where permissive licensing matters more than absolute benchmark scores.

Eingabe

$0.20 / 1M

Ausgabe

$0.80 / 1M

Kontext

256K tokens; OpenRouter providers $0.13-$0.20 in / $0.53-$0.80 out; free tier ended

Inkling

Thinking Machines Lab

Bedingt

Inkling ist der stärkste Grund bisher, Open-Weight-multimodale KI ernst zu nehmen. Es liefert wettbewerbsfähiges Reasoning und Coding mit erstklassiger Open-Weight-Sicherheit und kontrollierbarem Denkaufwand, alles unter Apache 2.0. Aber Thinking Machines ist ehrlich, dass es nicht das insgesamt stärkste ist — die Faktentreue hinkt hinterher (SimpleQA 43,9 %) und Self-Hosting erfordert ernsthafte Hardware (2 TB VRAM). Eingeschränkt: eine überzeugende Basis für Organisationen, die ein multimodales Modell besitzen und feintunen müssen, nicht für Teams, die Spitzenleistung von der Stange wollen.

Eingabe

$1.87 / 1M tokens

Ausgabe

$4.68 / 1M tokens

Kontext

Up to 1M tokens native; Tinker 64K and 256K tiers; rates at 50% limited-time discount (list $3.74 / $9.36)

Kimi K3

Moonshot AI

Bedingt

Moonshot AI's open-weight MoE model (Kimi K3 License, weights on Hugging Face) offers a 1M-token context at $3/$15 per 1M tokens and holds #1 on Arena WebDev (preliminary), but still trails Fable 5 and GPT-5.6 Sol on precise execution, and serving the weights needs 64+ accelerators. The rating stays conditional on legal risk: on July 22, 2026 the White House OSTP accused Moonshot of distilling Fable 5 and Treasury threatened sanctions. We have not re-verified that dispute's status this run.

Eingabe

$3.00 / 1M

Ausgabe

$15.00 / 1M

Kontext

1M ctx, $0.30 cached. Weights live (Jul 27).

Kolibri-1 is Aleph Alpha's English-German mixture-of-experts model with Apache 2.0 weights on Hugging Face, and a credible open-weight option for teams that must run German-language workloads on their own hardware. The catch is operations: the FP8 weights need about 78 GB of GPU memory, the pages we read name no hosted API or price, and every benchmark comes from Aleph Alpha's own harnesses. Right for public-sector and industrial teams that can self-host. Skip it for hosted-API needs or other languages.

Eingabe

Free (open weights)

Ausgabe

Free (open weights)

Kontext

262K native, 1M extrapolated; ~78 GB FP8

Laguna S 2.1

poolside

Bedingt

Laguna S 2.1 ist das erste glaubwürdige westliche Open-Weight-Coding-Modell seit 11 Monaten — ein 118B MoE, das nur 8B Parameter pro Token aktiviert und dennoch 70,2 % auf Terminal-Bench 2.1 erreicht und damit Modelle mit 10-facher Größe schlägt. Aber es ist nicht an der Frontier — GPT-5.6 Sol und Kimi K3 liegen weit vorne — und bekannte Einschränkungen umfassen Harness-Overfitting und JSON-Verstümmelung. Am besten für selbstgehostetes agentenbasiertes Coding, wo Datensouveränität wichtiger ist als absolute Benchmark-Spitzenwerte.

Eingabe

$0.10 / 1M

Ausgabe

$0.20 / 1M

Kontext

OpenMDW-1.1 license, 1M context

Llama 4 Maverick

Meta

Bedingt

Das fuehrende Open-Source-Modell fuer Teams, die selbst hosten wollen, Datenhoheit benoetigen oder API-Anbieterbindung vermeiden moechten.

Eingabe

Free (self-hosted) or $0.20 / 1M tokens (hosted)

Ausgabe

Free (self-hosted) or $0.60 / 1M tokens (hosted)

Kontext

1M tokens

MAI-Code-1-Flash

Microsoft

Bedingt

MAI-Code-1-Flash was Microsoft's first in-house coding model, a sparse MoE priced at $0.75/$4.50 per 1M tokens. GitHub retired it from Copilot on 10 September 2026 and names MAI-Code-1.1-Flash, at $0.20/$1.20, as the replacement; it no longer appears in Copilot's supported-models or pricing tables. Copilot users should switch to 1.1-Flash, which is cheaper and scores higher on Terminal-Bench 2.1.

Eingabe

$0.75 / 1M tokens

Ausgabe

$4.50 / 1M tokens

Kontext

Retired from Copilot Sep 10 2026; 256K ctx

MAI-Code-1.1-Flash

Microsoft

Bedingt

MAI-Code-1.1-Flash ist Microsofts aufgefrischtes Small-Tier-Coding-Modell für Copilot — ein Viertel der Kosten und 22 % besser bei Terminal-Bench 2.1 als sein Juni-Vorgänger, jetzt mit nativer Vision. Es ist für iterative Arbeit mit hohem Volumen gebaut statt für Frontier-Reasoning, was es zum günstigsten ernstzunehmenden Coding-Modell in Copilot macht. Greifen Sie danach für alltägliche Completions, Refactorings und bildgestütztes Debugging; verwenden Sie ein Flaggschiff für komplexe Architektur.

Eingabe

$0.20 / 1M

Ausgabe

$1.20 / 1M

Kontext

256K context

MAI-Cyber-1-Flash

Microsoft

Bedingt

Microsoft's first in-house cybersecurity model, built on the MAI-Thinking-1 family and embedded inside the MDASH multi-agent harness, handles about 90% of security workloads at roughly half the cost of prior configurations and reserves GPT-5.4 for the hardest 10%. Microsoft reports 96% on CyberGym by the any-crash metric and 86.3% by the stricter final-submission metric, both from a tuned full-system configuration rather than an independent model-versus-model test. Available only through Azure AI Foundry with vetted enterprise access; no standalone pricing or public API.

Eingabe

N/A (MDASH SCU consumption)

Ausgabe

N/A (MDASH SCU consumption)

Kontext

Azure AI Foundry vetted access

Conditional. Mercury 2.5 is Inception's diffusion LLM, and raw output speed is the reason to pick it: Artificial Analysis ranks it among the fastest models it measures, and it ships tunable reasoning, parallel tool calls and structured outputs. The deciding caveat is quality: Artificial Analysis places its intelligence just below the median, and Inception's parity claims name no benchmark. Suits latency-bound voice agents, RAG pipelines and coding subagents, not hard reasoning.

Eingabe

$0.20 / 1M tokens

Ausgabe

$0.75 / 1M tokens

Kontext

260K ctx; 80% off at launch: $0.04/$0.15

Mistral Large 4

mistral

Bedingt

Conditional: Mistral Large 4 is a public preview, not a finished release, since Mistral says its reinforcement learning run is still in flight and promises weights "by the end of the month". It is a mixture-of-experts model of about 1 trillion parameters, 52B active, with a 1M context on its docs page and a $1.36 input, $4.18 output list price per 1M tokens. Most benchmarks are Mistral's own. Trial it now if you want an open-weight-bound model, but wait for weights and licence before committing.

Eingabe

$1.36 / 1M tokens

Ausgabe

$4.18 / 1M tokens

Kontext

1M ctx; preview; sale $0.68/$2.09 on docs

Muse Glimmer

Meta

Bedingt

Muse Glimmer ist Metas 30B Open-Weight-Agentenmodell, destilliert aus dem geschlossenen Muse-Spark-Frontier-Modell und unter Apache 2.0 veröffentlicht. Mit 4-Bit-Quantisierung führt es mehrschrittiges Tool Calling, multimodales Reasoning und Fehlerbehebung auf einer einzelnen 24-GB-Consumer-GPU ohne Cloud-Abhängigkeiten aus.

Eingabe

Free (open weights)

Ausgabe

Free (open weights)

Kontext

131K tokens

Muse Spark 1.1

Meta

Bedingt

Metas erstes kostenpflichtiges KI-Modell und erster Vorstoß in den kommerziellen Coding-Markt — 1,25 $/4,25 $ pro 1M Token mit 1M Kontext, nativer Subagenten-Orchestrierung, MCP/Tool-Unterstützung und Computer-Use-Fähigkeit zu etwa 4× unter Anthropic/OpenAI-Preisen. Führt bei Tool-Use-Benchmarks, liegt aber bei reinem Coding 7 Punkte hinter Opus 4.8 auf SWE-Bench Pro. Closed-Weight, US-only öffentliche Vorschau. Am besten für Teams, die Tool-Use-Orchestrierung zu niedrigen Kosten priorisieren.

Eingabe

$1.25 / 1M tokens

Ausgabe

$4.25 / 1M tokens

Kontext

1M tokens

Muse Spark 1.2

Meta

Bedingt

Metas Coding-fokussiertes Modell-Update, co-trainiert mit dem Muse Code Harness — 82,9 % bei Terminal-Bench 2.1 und 59,3 % bei DeepSWE 1.1, nur von Claude Opus 5 bei Ersterem übertroffen. Standardpreise unverändert bei $1,25 Eingabe / $4,25 Ausgabe pro 1M Token.

Eingabe

$1.25 / 1M tokens

Ausgabe

$4.25 / 1M tokens

Kontext

1M tokens

Muse Spark 1.3

Meta

Bedingt

Muse Spark 1.3 is Meta's strongest coding and long-context model yet, and an easy drop-in for anyone already on 1.2: the standard token price is unchanged while software-engineering and million-token retrieval results move sharply up. The deciding caveat is that the headline agent numbers come from a max reasoning configuration Meta has not released, while the shipping tier runs slower and costs more per task. Right for cost-sensitive coding teams inside Muse Code; wrong for anyone who needs independently reproduced benchmarks.

Eingabe

$1.25 / 1M tokens

Ausgabe

$4.25 / 1M tokens

Kontext

1M ctx; contributor tier $0.10/$0.20

Muse Spark Open

Meta

Bedingt

Muse Spark Open is Meta's open-weight release of Muse Spark, an 816B mixture-of-experts model with 42B active parameters and a 1M-token context, available as downloads and from seven hosted providers. Conditional: the Muse Model License has limited commercial terms we did not read, the Meta pages we read give no hosted price, and Meta says its full evaluation report is still to come. It suits teams that want to self-host or shop between hosts, not buyers who need audited terms or independent benchmarks.

Eingabe

Free (open weights)

Ausgabe

Free (open weights)

Kontext

1M ctx; Muse Model License (limited commercial terms); no hosted price read

Muse Wonder

Meta

Bedingt

Muse Wonder is Meta's most capable model on its API, but it is still a preview and we could not read a single benchmark score behind the launch claims. Its list price is 3.2x Muse Spark 1.3 on input and 4.7x on output, and Meta says the full safety report is still to come. It suits teams already on Muse Spark who want to test it on long-horizon agent work, not buyers who need independently checked results.

Eingabe

$4.00 / 1M tokens

Ausgabe

$20.00 / 1M tokens

Kontext

1M ctx; preview; cached $0.20

Mythos 5 (Claude Mythos 5)

Anthropic

Bedingt

Das stärkste Cybersecurity-Modell der Welt (83,8 % CyberGym), aufgebaut auf den Gewichten von Fable 5, mit aufgehobenen Sicherheitsklassifikatoren. Der Zugang wurde nach der Exportdirektive vom 12. Juni ausgesetzt und am 26. Juni teilweise für ~100 designierte US-Organisationen kritischer Infrastruktur wiederhergestellt (Project Glasswing). Das Urteil bleibt CONDITIONAL: Die Fähigkeiten sind Elite, aber der Zugang ist an eine bundesbehördliche Designation gebunden. Keine öffentliche API: nutzbar nur, wenn Ihre Organisation auf der Liste des Handelsministeriums steht.

Eingabe

$10 / 1M tokens

Ausgabe

$50 / 1M tokens

Kontext

1M tokens (restricted access)

Ornith-1.0

DeepReinforce

Bedingt

Eine starke MIT-lizenzierte Open-Source-Modellfamilie für agentisches Coding, die Spitzen-Benchmarks erreicht — schlägt Claude Opus 4.7 sowohl auf Terminal-Bench 2.1 als auch SWE-Bench Verified — erfordert aber Self-Hosting mit erheblicher GPU-Infrastruktur und stammt von einem neuen Labor mit begrenzter Erfolgsbilanz.

Eingabe

Free (MIT license)

Ausgabe

Free (MIT license)

Kontext

256K tokens

Palmyra X6

Writer

Bedingt

Palmyra X6 ist Writers agentisches Flaggschiff-Modell für Marketing- und Revenue-Workflows; post-trainiert auf dem Open-Weights-Modell GLM-5.2 von Z.ai zu $2/$8 pro 1M Token, ein Bruchteil der Frontier-Preise. Jede hervorgehobene Kennzahl stammt aus Writer-eigenen Messungen, und die GLM-5.2-Herkunft ist eine echte Beschaffungsfrage. Nutzen Sie es innerhalb von Writer für kostensensible GTM-Agent-Arbeit; für unabhängig gebenchmarkte allgemeine Leistungsfähigkeit schauen Sie sich anderweitig um.

Eingabe

$2 / 1M

Ausgabe

$8 / 1M

Kontext

1M context, 8K max output

Qwen-Image-2.1

Alibaba

Bedingt

Qwen-Image-2.1 is Alibaba's open-weights model for both text-to-image generation and editing, with native transparent RGBA output and multi-image composition from up to 10 references. Conditional because the weights ship under the Qwen Research License, which allows non-commercial use only, and we found no readable benchmark scores. Right for researchers and hobbyists running image generation and editing locally through Diffusers.

Eingabe

Free (non-commercial weights)

Ausgabe

Free (non-commercial weights)

Kontext

Qwen Research License; non-commercial

Qwen3.8-27B

Alibaba

Bedingt

Qwen3.8-27B ist Alibabas Open-Weight-Modell mit 27 Milliarden Parametern, jetzt unter Apache 2.0 mit 262K nativem Kontext und Single-GPU-Deployment verfügbar. Es ist der selbst-hosting-fähige Begleiter des 2,4T-Flaggschiffs Qwen3.8-Max. Die Benchmarks bleiben Alibaba-eigene Zahlen, daher das Urteil 'conditional': ein echtes, lauffähiges Modell, dessen vom Anbieter behauptete Coding-Gewinne noch unabhängig reproduziert werden müssen.

Eingabe

Free (open weights — self-hosted)

Ausgabe

Free (open weights — self-hosted)

Kontext

262K native ctx, 1M extended; Apache 2.0

Qwen3.8-Omni-Flash

Alibaba

Bedingt

Qwen3.8-Omni-Flash takes text, images, audio and video in one call with a 1M-token context, and Alibaba says it cut the per-hour price of audio input by more than 98% against the previous version. Conditional because every benchmark we found is Alibaba's own and the model answers in text only; spoken replies need a separate model, such as the Realtime variant, at higher rates. Right for teams building audio-visual agents on Alibaba Cloud.

Eingabe

$0.15 / 1M

Ausgabe

$0.47 / 1M (text)

Kontext

1M ctx; cache hit $0.016/1M; International

Sakana Fugu

Sakana AI

Bedingt

Multi-Agent-Orchestrator, der Frontier-Modell-Niveau in zentralen Benchmarks durch Routing über einen Pool von Expertenagenten erreicht — aber intransparentes Routing und Plattformabhängigkeit schränken die Nachvollziehbarkeit ein.

Eingabe

$5 / 1M tokens

Ausgabe

$30 / 1M tokens

Kontext

272K tokens (>272K: $10/$45 per 1M)

SWE-2

Cognition

Bedingt

SWE-2 is Cognition's coding model for Devin, post-trained from Kimi K3, and Cognition claims it lands within one point of Fable 5.1 on FrontierCode 1.1 Main at 64% lower cost. Conditional because every SWE-2 score is vendor-reported, the same table shows it far behind Fable 5.1 and GPT-6 Astra on Terminal-Bench 4, and it runs only inside Devin. Right for Devin Desktop and CLI users; wrong for anyone who needs an API model or independently verified results.

Eingabe

N/A (no standalone API)

Ausgabe

N/A (no standalone API)

Kontext

Devin only; Enterprise $0.75 in/$3.75 out per 1M

Ternary Bonsai 2 27B

PrismML

Bedingt

Ternary Bonsai 2 27B squeezes Qwen3.8 27B into about 5.9GB at 1.76 effective bits, so a 262K-context text-and-image model fits on a Mac or a single consumer GPU under Apache 2.0. Conditional because the quality figures are PrismML's own, its blog and model card report different aggregate scores, and the files need PrismML's custom kernels: stock llama.cpp will not run them. Right for local-inference users short on memory.

Eingabe

Free (open weights)

Ausgabe

Free (open weights)

Kontext

262K ctx; self-hosted; PrismML kernels required

Gemini 3.5 Flash Cyber

Google

Vorsicht

Gemini 3.5 Flash Cyber ist Googles erstes Cybersicherheits-spezialisiertes Modell, feinabgestimmt aus 3.5 Flash, um Schwachstellen zu finden, zu validieren und zu patchen. Innerhalb der Multi-Agenten-Architektur von CodeMender erreicht es konkurrenzfähige CyberGym-Leistung gegenüber deutlich größeren Modellen wie Mythos — zu einem Bruchteil der Kosten. Doch die Einschränkungen sind gravierend: Es läuft ausschließlich innerhalb von CodeMender, ist nur für Regierungen und vertrauenswürdige Partner über ein limitiertes Pilotprogramm verfügbar, ohne öffentliche API, ohne veröffentlichte Benchmark-Zahlen und ohne Zeitplan für eine breitere Freigabe.

Eingabe

N/A

Ausgabe

N/A

Kontext

CodeMender-only, government-gated pilot

Gemini 3.5 Pro

Google

pending

Googles Frontier-Modell vor der allgemeinen Verfügbarkeit, inzwischen bei der vierten Verschiebung und laut Bloomberg „Monate hinter dem Zeitplan“. Der 2M-Token-Kontext, die native multimodale Pipeline und das Deep-Think-Reasoning bleiben reale Fähigkeiten, doch ein bestätigter GA-Zeitplan existiert nicht. Die Coding-Fähigkeit ist das zentrale Defizit.

Eingabe

Not yet published

Ausgabe

Not yet published

Kontext

Est. $5-15/$30-60 per 1M. 2M ctx. Not GA.

Gemini 4

Google

pending

Google DeepMinds bestätigtes Next-Gen-Modell. Pre-Training begann am 21. Juli 2026. Keine Benchmarks, keine Preise, keine Specs, kein GA-Termin.

Eingabe

$2 / 1M tokens (through Nov 30)

Ausgabe

$10 / 1M tokens (through Nov 30)

Kontext

$2/$10 to Nov 30, then $4/$20; paid tier only

Qwen3.8-Max

Alibaba

pending

Multimodales 2,4B-Parameter-MoE-Flaggschiff. Noch nicht GA: keine Pro-Token-API-Preise, keine Benchmark-Tabelle, keine Open Weights veröffentlicht. $6/Monat-Einstieg über Token Plan Lite.

Eingabe

$2.00 / 1M tokens

Ausgabe

$6.00 / 1M tokens

Kontext

Model Studio (Singapore); one tier up to 1M tokens